Macroeconomic Dynamics (2024), 28, pp.

612–646
doi:10.1017/S1365100523000184

A RT I C L E

Stock market alphas help predict macroeconomic

innovations
Mao-Wei Hung1 and Andy Jia-Yuh Yeh2 , 3
1 Graduate Institute of International Business, National Taiwan University, Taipei City, Taiwan
2 Graduate Institute of Finance, National Taiwan University, Taipei City, Taiwan
3 Brass Ring International Density Enterprise, Hong Kong

Corresponding author: Andy Jia-Yuh Yeh; Email: [email protected]

Abstract
We extract dynamic conditional factor premiums from the Fama-French factor model and find that most
anomalies disappear after one accounts for time variation in these premiums. Vector autoregression evi-
dence shows that mutual causation between dynamic conditional alphas and macroeconomic surprises
serves as a core qualifying condition for fundamental factor selection. This economic insight is an incre-
mental step toward drawing a distinction between rational risk and behavioral mispricing models. To the
extent that dynamic conditional alphas can reveal the marginal investor’s fundamental news and expecta-
tions about the cross-section of average asset returns, our economic insight helps enrich macroeconomic
asset return prediction.

Keywords: Fama-French multi-factor models; vector autoregressions; Granger causation tests; dynamic conditional alphas;
macroeconomic innovations; asset return anomalies

1. Introduction
Fama and French’s (1992, 1993, 1996, 1998, 2015, 2016) seminal contributions shed skeptical light
on the empirical performance of the Capital Asset Pricing Model (CAPM) of Sharpe (1964) and
Lintner (1965). The subsequent search for a better asset return model has been a hot pursuit for
many financial economists (Fama and French (1993, 1995, 1996, 1998, 2006b, 2008, 2015; Hou
et al. (2014)). Financial economists engage in the relentless debate over whether dramatic move-
ments in market valuation such as the Global Financial Crisis reflect a “rational” fair-value price
correction or a lack of compensation for risk. The debate has left many financial economists at a
time-worn impasse (Fama and French, 2004). Empiricists continue to discover new asset anoma-
lies (e.g., Titman et al. (2004); Fama and French (2006b, 2008); Cooper et al. (2008); Li et al. (2009);
Novy-Marx (2013)).
Kozak et al. (2017; 2018) recent studies show that many recent empirical horse races cannot
draw a distinction between both rational and behavioral theories of average return evolution.
Kozak et al. (2018) report that a factor model with a small number of statistical principal-
components (PCs) performs as well as the prior factor models (e.g. Fama and French (1993);
Hou et al. (2015); Fama and French (2015); Novy-Marx and Velikov (2016); Barillas and Shanken
(2018)). For typical portfolios, these few factors dominate the covariance matrix of returns. In the
absence of near-arbitrage investment opportunities, these factor models include only a few domi-
nant factors, which can be fundamental characteristics such as size and book-to-market or purely
statistical PCs. Because these models exhibit few conceptual links to investor beliefs and prefer-
ences, the q-theoretic factor models are as much “behavioral models” as the models are “rational

This research paper has been accepted for publication in Macroeconomic Dynamics, April 2023.

C The Author(s), 2023. Published by Cambridge University Press. This is an Open Access article, distributed under the terms of the
Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and
reproduction, provided the original article is properly cited.

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 613

models” (cf. Berk et al. (1999); Johnson (2002); Gomes et al. (2003); Liu et al. (2011); Liu and
Zhang (2008); Lin and Zhang (2013); Liu and Zhang (2014)). In a similar vein, the prior tests of
characteristics-versus- covariances inform us little of whether and how behavioral barriers such
as investor sentiment and overconfidence affect asset returns (e.g., Daniel and Titman (1997);
Brennan et al. (1998); Davis et al. (2000); Stambaugh and Yuan (2017)). Overall, it seems futile
to implement these horseraces to assess the complex convolution of both investor sentiment and
rationality in most asset return tests.
In the current study, we carry out an alternative approach to tackling this important issue in
finance. We first extract all dynamic conditional factor premiums from the Fama and French
(2015) model and then find that most anomalies disappear after one accounts for time variation
in these premiums. Granger-causality tests suggest new mutual causation between dynamic con-
ditional alpha spreads and macroeconomic innovations. This mutual causation therefore serves
as a core qualifying condition for factor selection with sound economic motivation. In the cur-
rent study, our economic insight is an incremental step toward drawing a distinction between
both rational risk and behavioral models in response to Kozak et al. (2017; 2018). To the extent
that macroeconomic surprises manifest in the form of dynamic conditional alphas, this causation
can reveal the marginal investor’s fundamental news and expectations about the cross-section of
average returns. Hence, our evidence contributes to macroeconomic asset return prediction.
Kozak et al. (2018) point out that the presence of arbitrageurs will connect the covariance
structure to the cross-section of average returns. Insofar as this cross-section at least partially
reflects investor sentiments such as overconfidence, salience of recent experience, and other cog-
nitive biases, behavioral mispricing factors can help price assets with reasonable bounds on the
Sharpe ratios. In this context, these behavioral factors load reasonable premiums to correctly price
assets. However, these behavioral factors and premiums “will not necessarily covary with aggre-
gate macroeconomic risks, as the [fundamental factors] would. . .” (cf. Kozak et al. (2017, 2018);
Daniel et al. (2017); Daniel et al. (2017)). In fact, Kozak et al. (2018) suggest several alternative
routes to design asset return tests that are more informative about investor beliefs, behaviors, and
preferences:
“To devise tests that are more informative about investor beliefs, researchers must exploit addi-
tional predictions of a factor model that relate returns to other data such as macroeconomic
variables, information on portfolio holdings, or data on investor beliefs” (cf. Kozak et al. (2018):
Section III.C first paragraph with our own bold italic emphasis).
In the current study, we pick the low-hanging fruit through an empirical analysis of mutual
causation between macroeconomic innovations and dynamic conditional factor premiums. To
the extent that macroeconomic innovations manifest in the form of dynamic conditional alphas
and betas, the conditional moments of returns and factors convey information about the cross-
section of average returns. This causation can thus serve as a core qualifying condition for valid
and sound factor selection in macroeconomic asset return prediction. In particular, our evi-
dence lends credence to the ubiquitous use of Fama and French (2015) factors that can reveal
the marginal investor’s response to fundamental expectations about the cross-section of aver-
age returns. At the same time, however, our econometric tests support Fama and French’s (1996,
2008, 2015, 2016) perennial reluctance to consider the Carhart (1997) momentum factor in their
factor model. We connect our dynamic conditional factor model results to recent advances in
the intertemporal CAPM context (Merton (1973); Campbell (1993); Campbell and Vuolteenaho
(2004); Campbell et al. (2017)). On balance, our current study represents an incremental step
toward better deciphering a distinction between the rational risk paradigm and the behavioral
mispricing conjecture.
We acknowledge the fact that some recent studies replicate a broader basket of anomalies
(cf. Fama and French (2016); Harvey et al (2016); Hou, et al. (2017); Harvey (2017); Chordia
et al. (2017)). Our primary and ultimate goal is not to compete with these prominent authors
with more empirical replication. Instead, we establish new Granger causation between dynamic

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 614 M.-W. Hung and A. J.-Y. Yeh

conditional alpha spreads and macroeconomic innovations as a core qualifying condition for fun-
damental factor selection in macro asset return prediction. This condition adds sound economic
rigor and intuition to macroeconomic asset return prediction. So this condition contributes to our
fresh insight that bilateral causation between macroeconomic surprises and dynamic conditional
alpha spreads reflects the marginal investor’s fundamental news and macro expectations about the
cross-section of average returns. This fresh insight can help demystify the empirical puzzle that
Kozak et al. (2018) suggest in their recent research.
Macroeconomic innovations move in tandem with dynamic conditional factor premiums that
provide unique economic insights into the implicit nexus between state-dependent alphas and
betas across fundamental factors. This nexus reveals rich information about the conditional fac-
tor covariance matrix in contrast to the unconditional counterpart, the latter of which omits
informative restrictions across the conditional moments of both asset returns and factors from
the empirical assessment of goodness-of-fit (Nagel and Singleton (2011)). Not only do dynamic
conditional alpha spreads change over time, but these alpha spreads also exhibit a robust causal
relation with macroeconomic surprises in a standard vector autoregressive system (Sims (1980)).
Macroeconomic innovations Granger-cause most dynamic conditional alpha spreads except for
momentum and partial value. Granger causation runs in a bilateral direction such that dynamic
conditional alpha spreads both lead and convey material information about macro innovations
(cf. our subsequent explanatory text on Section 4 and Tables 6, 7, 8 and 9).
This evidence enriches our key interpretation of the intertemporal CAPM that macroeconomic
gyrations both lead and vary with the conditional expectations of terminal payoffs in the marginal
investor’s intertemporal selection (Merton (1973); Campbell (1993); Fama (1996); Campbell and
Vuolteenaho (2004); Campbell et al. (2017)). Macro innovations manifest in the form of dynamic
conditional alpha spreads that persist as abnormal returns. Therefore, mutual Granger causation
between macro surprises and dynamic conditional alpha spreads becomes an informative piece of
evidence that we can apply to help resolve some long prevalent asset anomalies.
With this theoretical justification of the intertemporal CAPM, we contribute to the empirical
design of a workhorse asset return model by qualifying specific fundamental factors as useful
and reasonable state variables. Our core qualifying condition is equivalent to mutual causation
between macro surprises and alpha spreads that reflect changes in the conditional expectations
of structural shifts in terminal wealth for the marginal investor. To the extent that many residual
macroeconomic fluctuations lead dynamic conditional alpha spreads and vice versa, this causal
relation becomes a necessary condition for valid and effective factor selection in empirical asset
return research. For this pivotal purpose, sound theoretical justification of fundamental factors
with respect to causal macroeconomic innovations should precede pure empirical motivation in
macroeconomic asset return prediction (Harvey (2017); Harvey et al. (2016)).
Our current study also contributes to the conditional multifactor model literature. It is well-
known that a dynamic conditional mean-variance efficient return need not unconditionally price
the static portfolios with constant weights (cf. Cochrane (2005: 140)). Specifically, if a portfolio
return is on the conditional mean-variance efficient frontier, this return may or may not land on
the unconditional mean-variance efficient frontier. Cochrane (2005: 168) describes this issue in a
succinct statement: “Whether the [multifactor model] can be rescued by more careful treatment of
conditioning information remains an empirical question”. In the current study, we attempt to fill
this theoretical void by conditioning the main prediction of average returns on Fama and French
(2015) factors up to each time increment. Subsequent work substantiates our economic insight
that reconciles a reasonable array of anomalies within the dynamic conditional factor model.
Our unique use of fresh econometric tools serves as another empirical contribution to the
asset return literature. Both the conditional specification test and recursive multivariate filter
help extract dynamic conditional factor premiums that covary with macro surprises substantially
over time. Conditional on the expectation of zero prior prediction error, the recursive multi-
variate filter updates all of the dynamic alpha-beta factor premiums up to each time increment

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 615

in real-time. This important econometric innovation extends and generalizes the Fama-French
multiple-regression approach and therefore can become part of the standard toolkit for subse-
quent asset pricing analysis. This application reconciles a baseline array of anomalies with the
central theme of “dynamic” multifactor mean-variance efficiency (cf. dynamic MMVE in Fama
(1996) and Merton (1973)). Further, the concomitant tests provide evidence in support of this
notion. As a result, dynamic MMVE can serve as an informative benchmark for the empiri-
cal assessment of pervasive anomalies or portfolio strategies that generate persistent abnormal
returns in a static context. Overall, our dynamic conditional factor model thus has key implica-
tions for equity cost estimation, risk management, fund performance evaluation, and corporate
event assessment.
Applying the recursive multivariate filter adds value to the notion of dynamic multifactor
mean variance efficiency (MMVE) in the intertemporal asset pricing context (cf. Merton (1973);
Campbell (1993); Fama (1996); Campbell and Vuolteenaho (2004); Campbell et al. (2017)). In
this context, investors care about not only their terminal wealth but also state variables such as
human capital, labor income, consumption, and hedging investment opportunities that covary
with their terminal wealth (Fama and French (2004)). For instance, several studies suggest that
inter-industry heterogeneity in both human capital and labor mobility can help explain the cross-
section of average returns (Eiling (2013); Donangelo (2014)). An international factor model with
Epstein and Zin (1989) recursive investor preferences explains the high correlation of stock
market indices despite the low correlation of fundamental factors (Colacito and Croce (2011)).
A twin-country model can demystify both the carry trade puzzle and low correlation between
exchange rate movements and cross-country differences in total consumption in the intertempo-
ral context (Colacito and Croce (2013)). Hence, the Fama and French (2015) factors serve as valid
and relevant empirical hedging instruments for the marginal investor’s intertemporal selection
between his or her current and future investment opportunities. Our current work suggests that
the exclusion of Fama-French factors leads the econometrician to reject the null hypothesis of a
correct factor model specification (Fama and French, 2016). Specifically, we find mutual Granger
causation between macroeconomic surprises and dynamic conditional alphas for the Fama and
French (2015) fundamental factors, except for momentum and partial value. This latter falsifi-
cation provides empirical justification of Fama and French’s (1993, 1996, 2015, 2016) perennial
reluctance to encompass Carhart (1997) momentum as a new fundamental factor in the ratio-
nal risk paradigm. All of this evidence thus bolsters our intertemporal CAPM interpretation of
fundamental factors for the dynamic conditional factor model.
To the extent that macroeconomic surprises manifest in the form of dynamic conditional
alphas, this causation reveals the marginal investor’s fundamental news and expectations about
the cross-section of average returns. Our evidence enriches and contributes to the intertemporal
CAPM interpretation of dynamic conditional factor models. In this context, macro innovations
serve as fundamental news and expectations that induce cash-flow and future-risk betas as “bad
betas” or negative discount-rate betas as “good betas”. In accordance with this intertemporal
CAPM thesis, we expect assets with positive cash-flow shocks, future-risk spillovers, or subpar
discount-rate news to yield low average returns. Conversely, we would expect other assets with
negative cash-flow shocks, volatility declines, or optimistic discount-rate news to generate high
average returns. Therefore, mutual causation between macroeconomic innovations and dynamic
conditional alpha spreads serves as a core qualifying condition for fundamental factor selection
with sound economic rigor and intuition. This economic insight is one of our main contributions
to macroeconomic asset return prediction.
In addition to the use of a recursive multivariate filter for dynamic conditional alpha and
beta estimation, the conditional specification test helps draw a crucial distinction between both
the static and dynamic conditional factor models. This conditional specification test examines
whether the core distance between the static and dynamic conditional estimators turns out to be
significant so that there is sufficient evidence for one to reject the null hypothesis of a consistent

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 616 M.-W. Hung and A. J.-Y. Yeh

and efficient static specification. Under the alternative hypothesis, only the dynamic conditional
estimator is consistent although this more generic alternative specification may or may not be
efficient in the econometric sense. In our empirical analysis of 100 decile returns on the major
anomalies plus their respective 10 long-short stock portfolio strategies that focus on the extreme
deciles, about 95% of the stock portfolio tilts point to the statistically reliable rejection of the null
hypothesis that the static factor model is a correct specification. The key preponderance of empir-
ical results thus supports our chosen dynamic conditional factor model in contrast to the static
baseline factor model.
The remainder of our current study follows the structure below. Section 2 discusses the use
of a recursive multivariate filter for estimating dynamic conditional factor premiums from the
baseline Fama and French (2015) factor model. Section 3.1 describes the key datasets on the Fama-
French factors and anomalies. Section 3.2 discusses the core empirical evidence in support of
dynamic conditional factor premiums. Section 3.3 empirically analyzes each factor premium as
a typical financial time series that we extract from recursive multivariate filtration. Section 3.4
lists the conditional specification test evidence in favor of the dynamic conditional factor model.
Section 4 finds Granger causation between macroeconomic surprises and dynamic conditional
alpha spreads as the core qualifying condition for fundamental factor selection in modern asset
pricing model design. Section 5 concludes our study and offers new avenues for future research,
especially structural factor models with an economically intuitive and meaningful specification of
both investor beliefs and preferences.
The appendices offer supplementary evidence for our work. In particular, Appendix 1 helps
the reader visualize top-to-bottom-decile dynamic conditional alphas across several anomalies.
Appendix 2 presents the econometric test details for the canonical treatment of each dynamic
conditional factor premium as a unique typical financial time-series. Appendix 3 provides a list
of macroeconomic variable definitions and their data sources for our core vector autoregression
(VAR) empirical analysis of Granger bilateral causation between fundamental macro surprises
and dynamic conditional factor premiums. Appendix 4 encapsulates some elaborate discussions
on the conceptual nexus between our current study and several recent studies of empirical asset
return prediction. Appendix 5 presents the empirical results for dynamic conditional betas.

2. Methodology
In this section, we discuss our application of recursive multivariate filtration as an econometric
innovation. This filter helps extract major dynamic conditional factor premiums from Fama and
French’s (2015) factor model. We offer an intuitive explanation for connecting this filter to the
core notion of dynamic multifactor mean-variance efficiency (MMVE) (e.g., Merton (1973) and
Fama (1996)). Our intuitive explanation contributes to the empirical asset-pricing literature by
reconciling ubiquitous anomalies with dynamic multifactor portfolio efficiency.
An advantage of this unique econometric method is that we can assess whether the pervasive
asset pricing anomalies persist after one accounts for time variation in these dynamic conditional
factor premiums. We propose an alternative test of dynamic multifactor mean-variance efficiency
to complement Gibbons et al. (1989) F-test. Our central evidence suggests that the pervasive
anomalies of size, value, momentum, asset growth, operating profitability, and short-term and
long-term return reversals, are not robust after we account for the dynamic nature of conditional
factor premiums.1 A unique core implication of our empirical evidence is that dynamic MMVE
is essential to the design of a workhorse factor model for modern investment analysis. The prior
static factor models can be viewed as special cases of the more generalized dynamic conditional
factor model.
The vast majority of earlier studies of factor models rest upon the implicit assumption that
factor premiums are constant over time. Under this key assumption, the resultant static analysis
cannot account for the adverse effect of measurement noise that might be present in each state

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 617

variable. To the extent that conditional factor premiums vary over time, this measurement noise
can persist even in long-term data. Hence, the emergence and persistence of anomalous returns
may arise from the fact that the conventional static baseline model cannot adequately take into
account time variation in dynamic conditional factor premiums.
It is important for us to point out that our chosen use of a recursive multivariate filter differs
from the recent attempts by numerous proponents of the conditional CAPM or other conditional
factor model to allow each factor premium to change in short-window regressions, to move in
tandem with economic variables, or to co-vary in accordance with some specific structure of
autoregressive mean reversion (Lewellen and Nagel (2006); Fama and French (2006); Adrian and
Franzoni (2009); Ang and Kristensen (2012)).2 In contrast, the recursive multivariate filter can
allow conditional factor premiums to jointly covary in each time increment. This covariation is
not conditional on particular macroeconomic fluctuations. Neither does this covariation strictly
follow any arbitrary structure. As the recursive multivariate filter and conditional specification
test evidence both bolster the case for dynamic MMVE, these main results lend credence to the
empirical plausibility of dynamic conditional factor premiums. As dynamic conditional factor
premiums are highly volatile over time, this high volatility suggests a lack of statistical evidence
against the hypothesis that our chosen dynamic conditional factor model is correctly specified.3
Nagel and Singleton (2011) design a test of conditional moments of asset returns in a high-
dimensional context. It is well-known that it is more difficult to handle the multi-variate kernel
regressions as the number of dimensions increases in the Nagel and Singleton (2010) framework.
Thus, our chosen use of both recursive multivariate filtration and conditional specification test
evidence helps resolve this important issue in modern asset pricing model design. Not only do
we apply recursive multivariate filtration to extract informative time-varying conditional factor
premiums, but we also devise a new dynamic conditional specification test to empirically verify
factor premiums as dynamic financial time-series in the form of ARMA-EGARCH and ARMA-
GJR-GARCH stochastic processes. Appendix 1 provides the complete time-series visualization of
our dynamic conditional alphas over time.
The Fama and French (2015) five-factor model follows the canonical representation of equa-
tion (1) with static point estimates of factor premiums on the respective factors. This model
embeds the excess return on the CRSP value-weighted market portfolio. Each factor is the spread
between the average returns on the top 30% and bottom 30% stock deciles that the econometri-
cian sorts on size, book-to-market, asset growth, and operating profitability. Specifically, (Rkt –Rft )
and (Rmt –Rft ) denote the excess returns on the respective individual and market stock portfolios;
SMBt or Small-Minus-Big is the mean return spread between the top 30% and bottom 30% size
deciles; HMLt or High-Minus-Low is the mean return spread between the top 30% and bottom
30% book-to-market deciles; CMAt or Conservative-Minus-Aggressive equates the mean return
spread between the top 30% and bottom 30% investment deciles; and RMW t or Robust-Minus-
Weak is the average return spread between the top 30% and bottom 30% profitability deciles.
In comparison, we attempt to gauge the “dynamic estimates” of conditional factor premiums as
unique individual time-series in the alternative representation of equation (2). A primary com-
parison between equations (1) and (2) suggests that the former entails the static estimation of
point estimates of factor premiums, α, β m , β s , β h , β r , and β c , on the Fama and French (2015)
five-factors while the latter involves the dynamic estimation of time-series trajectories of condi-
tional factor premiums, α t , β mt , β st , β ht , β rt , and β ct , on the Fama and French (2015) five-factors.
For the practical purpose of our current analysis, the point estimates of static factor premiums in
equation (1) differ from the long-term mean values of dynamic conditional factor premiums in
equation (2) where these equations carry the Gaussian normal error terms εt and et :
 
Rkt − Rft = α + βm Rmt − Rft + βs SMBt + βh HMLt + βr RMWt + βc CMAt + εt (1)

 
Rkt − Rft = αt + βmt Rmt − Rft + βst SMBt + βht HMLt + βrt RMWt + βct CMAt + et (2)

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 618 M.-W. Hung and A. J.-Y. Yeh

Our recursive multivariate filter follows the dynamic multifactor representation below (Kalman
(1960); Harvey and Shephard (1993: 267-270); Lai and Xing (2008: 130-133); Tsay (2010: 591)):
β t+1 = At β t + ut+1 (3)

r t = F t β t + vt (4)
where β t is a (k + 1) × 1 vector of conditional factor premiums at each time increment; At is a
(k + 1) × (k + 1) identity matrix of linear dynamic variation in the state equation equation (3); r t
is a vector of excess returns on each portfolio; F t is a T × (k + 1) matrix of Fama-French factors
plus an intercept in the measurement equation equation (4); and ut and vt are independent ran-
dom vectors with E(ut ) = 0, cov(ut )= u , E(vt ) = 0, and cov(vt )= v . The dynamic states β t are
unobservable. The observations are the excess returns r t that are linear transformations of time-
varying factor premiums β t via the matrix F t plus the unobservable random disturbances ut . The
recursive multivariate filter is the recursive minimum-variance linear estimator of β t based on
the observations up to each time increment. We can define Pt|t-1 as the covariance estimator
of the unobservable state β t , as well as the filter for the previous state β t|t-1 . The gain matrix
follows the form κ t below:
 −1
κ t = At Pt|t−1 F Tt F t Pt|t−1 F Tt +  v (5)

For better exposition, we summarize the recursive formula for the filter in equations (6)–(9):
 
β̂ t+1|t = At β̂ t|t−1 + κ t r t − F t β̂ t|t−1 (6)

Pt+1|t = (At − κ t F t ) Pt|t−1 (At − κ t F t )T +  u + κ t  v κ Tt (7)

 −1  
β̂ t|t = β̂ t|t−1 + Pt|t−1 F t F t Pt|t−1 F Tt +  v r t − F t β̂ t|t−1 (8)
 −1
Pt|t = Pt|t−1 − Pt|t−1 F Tt F t Pt|t−1 F Tt +  v F t Pt|t−1 (9)

where we can initialize recursions at β 1|0 =E(β 1 ) and P1|0 =cov(β 1 ). Lai and Xing (2008: 130–133)
provide a complete derivation of the recursive formula for the filter. Due to its recursive nature,
the filter ensures that the measurement noise between the real-time state and its most up-to-date
dynamic estimator is nil on average (i.e. the expectation of the last term in equation (8) equates
zero).
To compare each stock portfolio to the market portfolio or the dynamic MMVE Q-portfolio,
we can compute the quadratic Sharp-ratio square as α T V -1 α where α is a vector of alphas on the
deciles and V is the residual variance-covariance matrix. In accordance with the prior treatment
of Gibbons, Ross, and Shanken’s F-test (1989) (Campbell et al. (1997: 192-193); Cochrane (2005:
230-233)), we present this test statistic in equation (10):
  T −1
T−N −1 α̂ V̂ α̂
GRS = ∼ F(N, T − N − 1) (10)
N 1 + SR2m
where k = 5 is the number of regressors in the Fama and French (2015) factor model; N = 10
denotes the number of deciles; and SRm denotes the Sharpe ratio for the mean-variance efficient
market portfolio in the Sharpe-Lintner CAPM. This GRS F-test examines whether a vector of aver-
age alphas is jointly zero in a static sense. Several comments can be made on the GRS F-test. First,
the test assumes away the fact that conditional factor premiums can exhibit substantial variation
over time. There can be non-trivial measurement noise in the estimation of unknown parame-
ters in the multifactor model (Merton (1980); Black (1986)). Second, the test only investigates a

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 619

static version of portfolio efficiency with a single mean-variance efficient portfolio (Markowitz
(1952); Merton (1973); Fama (1996); Fama and French (1996)). The test may thus call for some
adjustment in the more plausible case where several assets span the mean-variance space and
then address the hedging concerns for the investors who care about their intertemporal portfolio
choices. In this light, the optimal benchmark portfolio should include the relevant state variables
that yield the largest possible Sharpe ratio over a sufficiently long time horizon. Third, the test rests
on the implicit assumption that the alpha spread is constant over time. In this case, the GRS F-test
may reject the correctly-specified multifactor model more often than one otherwise would in a
dynamic context. The subsequent analysis demonstrates the opposite case that there is pervasive
time variation in conditional factor premiums. Each dynamic alpha spread varies substantially
at different historical junctures and exhibits the well-known properties of a financial time series.
To the extent that the GRS F-test cannot take into account dynamic heterogeneity in conditional
factor premiums, it is important for the econometrician to design a more suitable test of dynamic
portfolio efficiency.
When the econometrician applies the recursive multivariate filter to extract the time-series
of dynamic conditional factor premiums, alpha spreads should enter the GRS-equivalent F-test
formula. Over each time increment the recursion is an independent estimation of dynamic factor
premiums. This estimation makes use of all the return data up to the point in time. It is important
to assess the joint significance of α t over each time increment. For this reason, we have to adjust
the degrees of freedom for the numerator of the GRS-equivalent F-test statistic:
  T −1
T−N −1 α̂ V̂ α̂  
AGRS = ∼ F T − N − k, T − N − 1 (11)
T−N −k 1 + SRm 2

We dub equation (11) the adjusted-GRS F-test. Cochrane (2005: 230–235) offers a GMM-
equivalent χ 2 -test that yields robust consistent standard errors to safeguard against both het-
eroskedasticity and serial correlation. In a similar vein, we run the adjusted-GMM χ 2 -test to
account for the key dynamic nature of each conditional alpha spread:
  T −1
T−N −1 α̂ V̂ α̂  
AGMM = ∼ χ2 T − N − k (12)
1 1 + SRm2

Our chosen use of a recursive multivariate filter adds value to the notion of dynamic MMVE in
the intertemporal asset-pricing context of Merton (1973), Campbell (1993), and Fama (1996).
In this dynamic context, investors care about not only their terminal wealth but also invest-
ment opportunities that these investors expect to face before they achieve their terminal wealth.
These marginal investors consider how their current wealth might co-vary with several state
variables such as human capital, labor income, consumption, and hedging investment oppor-
tunities that remain available after the present period (Fama and French (2004)). For instance,
inter-industry heterogeneity in human capital and labor mobility affects the cross-section of aver-
age returns (Eiling (2013); Donangelo (2014)). In addition, an international factor model with
Epstein-Zin (1989) recursive investor preferences helps explain the high correlation of global
market indices despite the low correlation of fundamental factors (Colacito and Croce (2011)).
A similar cross-country model helps demystify the carry-trade puzzle and the low correlation
between exchange rate gyrations and international differences in aggregate consumption in an
intertemporal asset-pricing context (Colacito and Croce (2013)). In this light, the Fama and
French (2015) factors serve as valid and relevant empirical hedging instruments for the marginal
investor’s intertemporal substitution between the current and future investment opportunities. A
subsequent strand of falsification tests suggests that the main exclusion of any one of the Fama
and French (2015) factors would lead the econometrician to reject the null hypothesis of our cor-
rect dynamic conditional factor model specification. All of this evidence thus shines fresh light
on a dynamic conditional interpretation of economic intuition behind the intertemporal CAPM

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 620 M.-W. Hung and A. J.-Y. Yeh

(Merton (1973); Campbell (1993); Fama (1996); Campbell and Vuolteenaho (2004); Campbell
et al. (2017)).
Fama and French’s (2015) five-factor model includes a unique set of fundamental factors (e.g.
Fama and French (1993, 1995, 1996, 1998, 2006b, 2012, 2015); Vassalou and Xing (2004); Petkova
(2006)). Our empirical study extends their “static” model to encapsulate time variation in condi-
tional factor premiums. This time variation suggests that each dynamic alpha oscillates too much
around nil so that the pervasive anomalies vanish. Once the econometrician accounts for the
dynamic nature of conditional factor premiums, each alpha spread between the extreme deciles
eventually disappears. A pivotal comparison hence has to be made against the dynamic MMVE
portfolio. For easier exposition, we dub this dynamic MMVE portfolio the “Q-portfolio” that
generates the largest possible average returns for each given set of asset return covariances and
variances with the valid fundamental factors (Fama, 1996). Without reinventing the wheel, we
use the Fama and French (2015) five-factors as state variables in our dynamic conditional fac-
tor analysis. Moreover, we propose an alternative Q-test of dynamic mean-variance efficiency by
comparing the Sharpe ratios for each long-short decile alpha-spread and the MMVE Q-portfolio.
A reasonable choice is to assume each Sharpe ratio to follow an independent normal distribution,
then the Q-test statistic conforms to the χ 2 distribution:
  −1
T−N −1 α̂ T V̂ α̂  
Q= −1
∼ χ2 T − N − k (13)
1 1 + q̂T Ŵ q̂
where q denotes a vector of Fama-French return spreads and W is the variance-covariance of
these return spreads. In effect, our chosen omnibus Q-test verifies whether the Sharpe ratio for
a particular portfolio tilt is so large that dynamic conditional alphas cannot be readily explained
by the Fama and French (2015) factor model. The Q-test statistic is asymptotically analogous to
the Wald test statistic of Ang and Kristensen (2012: 138–139) except here we regard the dynamic
MMVE Q-portfolio as the benchmark portfolio. In this latter case, the Q-test helps measure the
wedge between the Sharpe-ratio squares for the Q-portfolio and each stock portfolio tilt. In our
subsequent analysis, we use the Q-test to complement the AGRS F-test and the AGMM χ 2 -test
(aka the AGMM C-test).

3. Evidence
3.1. Data description
We retrieve the U.S. stock portfolio return data from Professor Ken French’s online data library.
The monthly stock dataset spans the 50-year period from January 1964 to December 2013. For
applying the recursive multivariate filter to this dataset, we run the filter on a training period
of 60 months. This technical choice allows the filter to adapt dynamic conditional factor premi-
ums by recursively learning from a sufficient set of prior information. The Fama-French factors
are the market risk premium (MRP), the return spread between the small-versus-big size portfo-
lios (SMB), the return spread between the high-versus-low book-to-market portfolios (HML), the
return spread between the robust-versus-weak portfolios in terms of their operating profitability
(RMW), and the return spread between the conservative-and-aggressive asset-growth portfolios
(CMA). For the practical purposes of this empirical analysis, we consider the value-weighted
deciles for the pervasive portfolio sorts of size, value (i.e. book-to-market, cash-flow-to-price,
dividend-to-price, and earnings-to-price), momentum, asset growth, operating profitability, and
short-term and long-term return reversals. For each of these sorts, we apply the recursive multi-
variate filter to extract dynamic conditional factor premiums on the Fama and French (2015)
factors and intercept term. This dynamic factor analysis focuses on whether the alpha spread
between the extreme deciles for each of the pervasive portfolio sorts is sufficiently large for us
to reject the null hypothesis of a correct factor model specification.

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 621

Table 1. Descriptive statistics for various stock return spreads

Stock return spread Mean Stdev Skew Kurt Min Med Max

Market risk premium (MRP) 0.49 4.50 −0.53 1.86 −23.24 0.86 16.10
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Small-minus-big (SMB) 0.31 2.92 0.37 3.50 −14.32 0.24 18.05

.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

High-minus-low (HML) 0.29 2.17 −0.14 3.26 −10.85 0.25 9.82

.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Robust-minus-weak (RMW) 0.25 1.52 −0.83 11.64 −12.76 0.26 7.60

.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Conservative-minus-aggressive (CMA) 0.15 1.18 0.32 1.08 −3.54 0.10 5.10

Size −0.37 4.87 −0.73 4.17 −32.21 −0.16 21.10

.......................................................................................................................................................................... ..................................................................................

Momentum 1.32 7.01 −1.49 8.04 −45.89 1.67 26.18

.......................................................................................................................................................................... ..................................................................................

Book-to-market 0.53 4.67 0.54 2.31 −13.58 0.44 26.73

.......................................................................................................................................................................... ..................................................................................

Cashflow-to-price 0.49 4.22 0.03 1.74 −20.27 0.39 16.02

.......................................................................................................................................................................... ..................................................................................

Dividend-to-price 0.08 5.38 0.03 2.55 −26.22 0.27 22.11

.......................................................................................................................................................................... ..................................................................................

Earnings-to-price 0.47 4.30 −0.02 1.57 −20.13 0.54 16.75

.......................................................................................................................................................................... ..................................................................................

Investment −0.48 3.26 −0.32 1.22 −15.47 −0.48 10.12

.......................................................................................................................................................................... ..................................................................................

Profitability 0.19 3.96 0.23 2.93 −19.68 0.35 22.46

.......................................................................................................................................................................... ..................................................................................

Short-term return reversal −0.36 5.35 −0.24 3.99 −26.65 −0.30 25.23
.......................................................................................................................................................................... ..................................................................................

Long-term return reversal −0.48 5.07 −0.98 4.67 −33.79 −0.16 18.94
This table summarizes the descriptive statistics for the various stock portfolio tilts. The first panel encapsulates the summary statistics for
the Fama and French (2015) return spreads such as the excess return on the CRSP value-weighted market portfolio (MRP), the return spread
between the top 30% small and bottom 30% big stocks (SMB), the return spread between the top 30% high book-to-market and bottom
30% low book-to-market stocks (HML), the return spread between the top 30% robust and bottom 30% weak stocks in terms of their relative
profitability (RMW), and the return spread between the top 30% conservative investment and bottom 30% aggressive investment stocks (CMA)
in developing the multifactor mean-variance efficient (MMVE) tangency portfolio. The second panel sums up the descriptive statistics for the
long-short trading strategy that involves both a long position in the top decile and a short position in the bottom decile for the pervasive asset
pricing anomalies such as size, value (book-to-market, cashflow-to-price, dividend-to-price, and earnings-to-price), momentum, investment,
profitability, short-term return reversal, and long-term return reversal. The descriptive statistics include each return spread’s mean value,
standard deviation, skewness, kurtosis, minimum, median, and maximum.

Table 1 shows the descriptive statistics for the Fama-French factors and return spreads that gen-
erate anomalous patterns in several prior studies (e.g. Fama and French (1996, 2015, 2016)). For
the size, asset investment growth, short-run return reversal, and long-run reversal sorts, the port-
folio strategy that involves both a long position in the top decile and a short position in the bottom
decile produces a negative average return spread. This evidence echoes the descriptive statistics
for the other decile sorts that yield positive mean return spreads. Because these time-series are
leptokurtic and exhibit fat tails, it is reasonable to conjecture that each factor premium may be
similar to a financial time series in the dynamic conditional factor context. Specifically, condi-
tional factor premiums vary much over time and exhibit pervasive autoregressive patterns in the
conditional mean specification and volatility clusters and asymmetries in the conditional variance
specification. Subsequent analysis provides a deeper exploration of these new patterns.
Table 2 lists the Sharpe ratios for the market benchmark portfolio, the Fama-French MMVE
Q-portfolio, and the portfolio sorts of size, value, momentum, profitability, investment growth,
and short-term and long-term return reversals. While the market portfolio attains a Sharpe ratio
of 0.1006, the MMVE Q-portfolio achieves a superior Sharpe ratio of 0.3006. The vast majority
of portfolio sorts generate Sharpe ratios that land within these bounds. A notable exception is
the short-term return reversal sort. The best Sharpe-ratio performers are the portfolio strategies
that exploit the return spreads between the top and bottom deciles of momentum, dividend-to-
price, asset growth, and short-term reversal with the respective Sharpe ratios of 0.2574, 0.2935,
0.2551, and 0.3362. In light of this evidence, most of the anomalies offer greater rewards that are
commensurate with their exposure to systematic risk in comparison to the CAPM. However, the
results also suggest that the anomalies largely lead to smaller mean excess returns per unit of risk

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 622 M.-W. Hung and A. J.-Y. Yeh

Table 2. Sharpe ratios for the market portfolio, the

Q-portfolio, and the anomalies

Stock portfolio sort Sharpe ratio

Market risk premium 0.1006

.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Multifactor MVE Q-portfolio 0.3006

.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Size 0.2084
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Momentum 0.2574
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Book-to-market 0.2016
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Cashflow-to-price 0.2172
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Dividend-to-price 0.2935
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Earnings-to-price 0.1358
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Investment 0.2551
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Profitability 0.1918
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Short-term return reversal 0.3362

.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Long-term return reversal 0.2271

This table summarizes the long-term Sharpe ratio for each stock
portfolio strategy in the period from January 1964 to December
2013. The Sharpe ratio is the ratio of excess stock return to
its standard deviation. The basket of stock portfolio strategies
encompasses the CRSP value-weighted market portfolio, the mul-
tifactor mean-variance efficient (MMVE) tangency stock portfolio
from the Fama and French (2015) joint return spreads for size,
value, investment, and profitability, as well as the ubiquitous asset
pricing anomalies such as size, value (book-to-market, cashflow-
to-price, dividend-to-price, and earnings-to-price), momentum,
investment, profitability, short-term return reversal, and long-term
return reversal. The Sharpe ratios land in the range of 0.1006 for the
CRSP value-weighted market portfolio to 0.3006 for the multifactor
MVE tangency portfolio and 0.3362 for the short-term return rever-
sal strategy. The other Sharpe ratios land within this intermediate
range.

relative to the MMVE Q-portfolio. Table 2 thus resonates with the Q-test evidence below that the
relative distance between the Sharpe-ratio squares for the dynamic MMVE Q-portfolio and each
of the stock portfolio tilts is not sufficiently large for one to reject the null hypothesis of a correct
dynamic conditional factor model specification.

3.2 Time-varying dynamic conditional alphas

Table 3 presents the time-varying Fama and French (2015) alphas across the deciles for each of the
portfolio sorts, t-tests of these alpha spreads between the extreme deciles with the Newey-West
(1987) standard-error correction that safeguards against potential heteroskedasticity and serial
correlation, F-tests of mean-variance efficiency (Gibbons et al. (1989)), and Q-tests that we pro-
pose as the appropriate test of dynamic portfolio efficiency to complement the filter. A first glance
at Table 3 indicates that most dynamic conditional alphas are statistically close to nil. Out of these
portfolio tilts, only the momentum, short-term reversal, and long-term reversal tilts yield signif-
icant alpha spreads between the top and bottom deciles in the range of 0.989, –1.180, and 0.426
(p-values < 0.001). Yet, the positive sign of the average alpha spread for long-term return reversal
is counter-intuitive and therefore instead suggests long-term return momentum. This evidence
contradicts the prior studies in support of long-term return reversal that can arise from the typi-
cal investor’s naïve extrapolation of past superior stock performance (DeBondt and Thaler (1985);
Lakonishok et al. (1994); Fama and French (1996)). With respect to return momentum (Jegadeesh
and Titman (1993, 2001); Chan et al. (1996)), the long-term average dynamic conditional alpha is
significant only for the extreme deciles. In this case, the top and bottom deciles produce significant

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 623

average conditional alphas of –0.569 and 0.42 respectively (p-values < 0.005). The resultant alpha
spread is therefore significant at the conventional statistical confidence level. Also, the dynamic
conditional alpha spread between the extreme short-term reversal deciles is significant in econo-
metric terms (p-value < 0.001). Whether these results are a statistical aberration calls for more
formal hypothesis tests on the short-term reversal and momentum phenomena.
In Table 3, all the AGRS F-tests, the AGMM C-tests, and the Q-tests unanimously suggest
that dynamic conditional alphas are jointly indistinguishable from zero for all the portfolio tilts.
The p-values are substantially near unity across the board. Hence, there is minimal evidence in
support of the alternative hypothesis that our dynamic factor model is incorrectly specified. The
main economic intuition is that the wedge between the Sharpe-ratio squares for the benchmark
portfolio and the long-short decile strategy is not large enough to justify the statistical rejection of
a dynamic variant of the Fama and French (2015) factor model. The conditional factor premiums
are too volatile for the econometrician to affirm the consistent outperformance of each portfolio
tilt once he or she takes into account the dynamic nature of these conditional factor premiums.
In Appendix 1, the time-series visualization of dynamic conditional alpha spreads between both
the top and bottom deciles corroborates this empirical fact. A falsification test suggests that the
highest F-test, C-test, and Q-test p-value is 0.04 (not shown in the tables and charts) when we
exclude any one of the Fama and French (2015) explanatory factors from the recursive estimation.
This falsification test evidence supports the joint insignificant of dynamic conditional alphas.
Harvey et al. (2016) introduce a multiple testing framework (e.g. Harvey and Liu (2014a,
2014b, 2014c, 2014d)) and provide a unique variety of historical significance cut-offs from the
first empirical tests in the 1960s to the present. This new strand of investment literature suggests
that we should raise the test hurdle substantially from a t-ratio of 2.0 to a t-ratio of 3.0 for most
cross-sectional asset-pricing tests. Specifically, Harvey et al. (2016) find that this higher hurdle
reduces the number of cross-sectional anomalies from 316 to only 2 that is, value and momentum
(cf. Asness et al. (2013); Fama and French (2016); Hou et al. (2017)). In addition, Harvey et al.
(2016) propose that a theoretically-derived factor should have a lower hurdle than an empirically-
discovered factor. Their central thesis suggests that a factor can be important in some economic
environments but unimportant in some other environments.
While our econometric innovation complements Harvey et al. (2016) multiple testing analysis,
our work serves as a time-series equivalent to their cross-sectional adjustment for asset-pricing
tests. Back-of-the-envelope calculations show that the typical stock portfolio’s Sharpe ratio has
to increase by at least 3 to 8.2 times for most dynamic conditional alphas to be jointly signif-
icant at the conventional confidence level. The critical values for the χ 2 -test with 525 degrees
of freedom are 603.31, 579.43, and 566.91 at the respective 99%, 95%, and 90% confidence lev-
els. Table 3 demonstrates that the highest C-test or Q-test statistic is 59.29 while the lowest
C-test or Q-test statistic is 8.97. Therefore, the smallest Sharpe ratio multiplier can be calculated
as (566.932/59.29)1/2 = 3.092 while the largest Sharpe ratio multiplier can then be calculated as
(603.31/8.97)1/2 = 8.201. As a result, the econometrician has to specify a higher test hurdle for
each anomaly. Across the deciles, most dynamic conditional alphas need to be larger on average
with significantly less variability for the Sharpe ratio to increase by at least 3 to 8 times. The equiva-
lent Sharpe ratio would be in the approximate range of 1.15 to 2.4 (cf. Kozak et al. (2017)). In other
words, our unique dynamic analysis of conditional factor premiums proposes raising the bar for
the econometric asset pricing test. This recommendation echoes the cross-sectional counterpart
of Harvey et al. (2016).

3.3 ARMA-GARCH representation of each dynamic conditional factor premium

In this section, we demonstrate that each dynamic conditional factor premium can be modeled
as a financial time-series. We apply both ARMA(1,1)-EGARCH(1,1,1) and ARMA(1,1)-GJR-
GARCH(1,1,1) conditional-mean-and-variance models to fit each conditional factor premium

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

624
Table 3. Average dynamic alphas, dynamic alpha spreads, NW t-tests, AGRS F-tests, AGMM χ 2 tests, and MMVE Q-tests

M.-W. Hung and A. J.-Y. Yeh

Portfolio Low Decile 2 Decile 3 Decile 4 Decile 5 Decile 6 Decile 7 Decile 8 Decile 9 High Spread F-test C-test Q-test

Size
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Alpha (test statistic) −0.076 −0.112 0.001 −0.092 0.080 0.104 0.090 0.104 0.053 −0.014 0.062 0.043 22.799 21.121
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.438 0.101 0.983 0.070 0.185 0.258 0.216 0.080 0.344 0.659 0.505 0.999 0.999 0.999
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Momentum
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Alpha (test statistic) −0.569 −0.171 0.016 −0.135 0.063 0.008 −0.021 0.039 0.164 0.420 0.989 0.051 34.776 32.216
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.005 0.215 0.857 0.072 0.392 0.879 0.722 0.635 0.038 0.001 0.000 0.999 0.999 0.999
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Book-to-market
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Alpha (test statistic) −0.001 0.067 −0.043 0.136 0.015 −0.126 −0.163 0.052 0.128 −0.042 −0.041 0.041 21.318 19.749
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.988 0.446 0.518 0.093 0.877 0.065 0.008 0.374 0.084 0.664 0.752 0.999 0.999 0.999
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Cashflow-to-price
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Alpha (test statistic) 0.050 −0.025 0.106 −0.064 −0.012 −0.012 −0.014 −0.226 −0.021 0.195 0.144 0.038 24.746 22.924
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.611 0.744 0.155 0.331 0.906 0.855 0.891 0.012 0.820 0.102 0.385 0.999 0.999 0.999
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Dividend-to-price
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Alpha (test statistic) 0.031 0.182 0.269 0.193 −0.231 0.015 −0.004 −0.049 −0.116 0.080 0.049 0.070 45.204 41.877
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.720 0.005 0.009 0.018 0.022 0.851 0.962 0.513 0.279 0.541 0.790 0.999 0.999 0.999
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Earnings-to-price
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Alpha (test statistic) 0.100 0.113 0.060 −0.020 0.048 −0.027 0.054 −0.032 −0.122 0.099 −0.001 0.016 9.683 8.970
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.256 0.048 0.472 0.803 0.610 0.783 0.434 0.709 0.052 0.396 0.995 0.999 0.999 0.999
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Investment
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Alpha (test statistic) −0.229 0.082 −0.104 −0.065 −0.005 −0.008 0.206 0.041 0.137 −0.086 0.143 0.049 34.142 31.629
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.029 0.223 0.187 0.307 0.948 0.888 0.044 0.358 0.039 0.328 0.310 0.999 0.999 0.999
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Profitability
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Alpha (test statistic) −0.108 0.122 −0.050 0.036 0.039 −0.148 0.045 −0.069 0.105 0.110 0.219 0.031 19.307 17.886
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.218 0.043 0.571 0.630 0.556 0.115 0.490 0.252 0.044 0.100 0.088 0.999 0.999 0.999
https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

Table 3. Continued

Portfolio Low Decile 2 Decile 3 Decile 4 Decile 5 Decile 6 Decile 7 Decile 8 Decile 9 High Spread F-test C-test Q-test

Short-term reversal
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Alpha (test statistic) 0.565 0.571 0.435 0.202 0.156 −0.147 −0.114 −0.262 −0.357 −0.615 −1.180 0.100 59.290 54.927
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.000 0.000 0.000 0.005 0.015 0.086 0.051 0.001 0.000 0.000 0.000 0.999 0.999 0.999
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Long-term reversal
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Alpha (test statistic) −0.339 0.034 −0.214 −0.124 −0.112 −0.055 0.100 0.254 0.144 0.088 0.426 0.052 27.063 25.071
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.003 0.748 0.019 0.195 0.040 0.400 0.152 0.000 0.018 0.345 0.001 0.999 0.999 0.999
Over the 50-year period from January 1964 to December 2013, the econometrician applies the recursive multivariate Filter to extract dynamic factor premiums from the Fama and French (2015) five-factor asset
pricing model. At each time increment, the econometrician takes into account the Fama and French (2015) factors such as the excess return on the market portfolio (MRP), the return spread between the top
30% small and bottom 30% big stocks (SMB), the return spread between the top 30% high book-to-market and bottom 30% low book-to-market stocks (HML), the return spread between the top 30% robust and
bottom 30% weak stocks in terms of their relative profitability (RMW), and the return spread between the top 30% conservative investment and bottom 30% aggressive investment stocks (CMA) to explain the
variation in the excess return on each stock decile for size, momentum, value (cf. book-to-market, cashflow-to-price, dividend-to-price, and earnings-to-price), investment, profitability, short-term return reversal,
and long-term return reversal. The econometrician presents the mathematical time-series representation below:

Rkt − Rft = αt + βmt (Rmt − Rft ) + βst SMBt + βht HMLt + βrt RMWt + βct CMAt + εt

Table 3 sums up the long-run average alpha for each stock decile sorted on size, value, momentum, investment, profitability, short-term return reversal, and long-term return reversal. The first 10 columns

Macroeconomic Dynamics
summarize each long-run average alpha and its corresponding p-value for the null hypothesis of zero dynamic alpha. The next column encapsulates the long-run average alpha spread for the long-short trading
strategy that involves both a long position in the top decile and a short position in the bottom decile throughout the 50-year period from January 1964 to December 2013. The last three columns summarize the
Gibbons et al. (1989) AGRS F-test, AGMM C-test, and AGMM Q-test results on each long-short trading strategy across the ubiquitous asset pricing anomalies such as size, value, momentum, investment, profitability,
short-term return reversal, and long-term return reversal. This evidence reports each test statistic and its corresponding p-value. For each hypothesis test, the econometrician applies the Newey and West (1987)
method with quadratic spectral kernel estimation to correct the standard errors to safeguard against any serial correlation and heteroskedasticity. The appendix depicts the time-series dynamic alphas for each
of the stock portfolio tilts that yield anomalous excess returns in static asset pricing analysis (i.e. size, value, momentum, investment, profitability, short-term return reversal, and long-term return reversal).

625
 626 M.-W. Hung and A. J.-Y. Yeh

that the econometrician extracts from a dynamic variant of the Fama and French (2015) mul-
tifactor model. Although it is possible to identify a better time-series representation for each
conditional factor premium, our goal here is more straight-forward. In fact, our primary objec-
tive is to apply the standard toolkit in time-series econometrics to establish the empirical fact that
each dynamic alpha or beta spread exhibits the major properties of most financial time series.
Each conditional factor premium embeds autoregressive mean reversion in the conditional mean
specification of ARMA(1,1), and volatility clusters and asymmetries in the conditional volatil-
ity specification of EGARCH(1,1,1) or GJR GARCH(1,1,1) (cf. Engle (1982); Bollerslev (1986);
Nelson (1991); Glosten et al. (1993)):

ARMA(1,1) conditional mean specification

mt = a + bmt−1 + cwt−1 + wt (14)

wt = h t εt (15)

EGARCH(1,1,1) and GJR-GARCH (1,1,1) conditional variance specifications

 
wt wt E(wt )
ht = exp d + e + f ln ht−1 + g − (16)
ht−1 ht−1 ht−1

2 2
ht = d + ewt−1 + fht−1 + gDt−1 wt−1 (17)
where mt is the dynamic conditional alpha or beta spread; wt is the residual error term; ht is the
conditional variance process; ε t is a Gaussian white noise; Dt denotes a binary variable with a
numerical value of unity if wt is negative or zero if wt is positive; a, b, c, d, e, f , and g are the
parameters for quasi-maximum likelihood estimation. The canonical ARMA model serves as the
conditional mean specification to capture any autoregressive mean reversion in the dynamic con-
ditional alpha or beta spread between the extreme deciles, while EGARCH or GJR-GARCH fits
the conditional variance specification to encapsulate any volatility clusters and asymmetries in the
current factor premium time-series under study.
It is important to draw a distinction between this time-series analysis and the prior studies of
multiple conditional factor models (Ferson and Harvey (1991, 1999); Fama and French (2006);
Ang and Chen (2007); Ang and Kristensen (2012)). In the current study, we need not impose any
a priori assumption about the dynamic evolution of conditional alpha or beta spreads, whereas,
the earlier studies of conditional factor models make specific assumptions about the time-series
behaviors of dynamic conditional factor premiums (such as structural breaks in autoregressive
mean reversion). Yet, the econometrician can readily fit an ARMA-EGARCH or ARMA-GJR-
GARCH model to characterize the dynamic evolution of each conditional alpha or beta spread
over time. This characterization entails both reasonable and flexible assumptions about the true
conditional mean and variance processes for each dynamic conditional factor premium.
This time-series analysis also differs from several earlier studies that exclusively focus on the
CAPM (cf. Adrian and Franzoni (2009); Ang and Chen (2007); Lewellen and Nagel (2006)). The
recursive multivariate filter helps extract dynamic conditional alphas and betas from the Fama and
French (2015) multifactor model, and then the econometrician can apply equations (14)–(17) to
model each dynamic conditional alpha or beta spread as a typical financial time series. While it is
reasonable to identify the “best” ARMA-GARCH representation for each conditional alpha or beta
spread, we aim to establish the empirical fact that each conditional alpha or beta spread exhibits
most prevalent properties of a typical financial time series. In turn, this empirical fact defies the
conventional wisdom of point estimates of factor premiums in most static time-series ordinary

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 627

least-squares regressions. Table 4 summarizes the empirical results in Appendix 2 (cf. Tables A2.1
to A2.6).
We summarize several bullet points from Appendix 2 and the tabular results therein:

1. These results allow us to establish the empirical fact that almost all the dynamic conditional
factor premiums exhibit the key properties of most financial time-series. Specifically, these
dynamic conditional alpha and beta spreads exhibit autoregressive mean reversion in the
conditional mean specification, and volatility clusters and asymmetries in the conditional
variance specification. Therefore, the conditional moments of factors and returns manifest
in the form of state-dependent alphas and betas, or dynamic conditional factor premiums,
across Fama and French’s (2015; 2016) fundamental factors (Nagel and Singleton (2011)).
Our subsequent analysis suggests bilateral causation between macroeconomic surprises
and conditional alpha spreads. Overall, the evidence enriches our chosen interpretation of
the intertemporal CAPM that most macroeconomic gyrations both lead and covary with
the conditional expectations of terminal wealth in the investor’s investment opportunity
set (cf. Merton (1973); Campbell (1993); Fama (1996); Campbell and Vuolteenaho (2004),
Campbell et al. (2009); Campbell et al. (2017)). To the extent that macroeconomic shocks
manifest in the form of persistent dynamic conditional alpha spreads, mutual causation
between macro surprises and alpha spreads hence becomes an informative piece of evi-
dence that we can exploit in order to resolve at least some of the prevalent abnormal returns
or stock market anomalies.
2. To the extent that stock market information serves as a useful indicator of macro
surprises, each dynamic conditional factor premium conveys rich information about
macroeconomic growth, market valuation, financial stress, cyclical variation, or forecast
combination. This inference calls for more corroboration in the core spirit of several recent
studies (Liew and Vassalou (2000); Vassalou (2003); Vassalou and Xing (2004); Petkova
(2006); Campbell et al. (2009); Campbell et al. (2017)).
3. Most of the average conditional alpha spreads are insignificant while the exceptions are
momentum and short-term reversal (with absolute t-ratios more than 2.9). For the latter
portfolio tilts, the respective conditional average alpha spreads are 1.03 and –1.12. These
conditional mean alpha spreads are close to the corresponding average dynamic condi-
tional alpha spreads for momentum and short-term return reversal of 0.989 and –1.18
in Table 3. Although these average alpha spreads seem to persist in the extreme deciles (cf.
Fama and French (2008; 2016)), it is key to recall the more formal Sharpe ratio test evidence
that the average alphas do not jointly differ from nil across all the momentum and short-
term return reversal deciles. In other words, these dynamic conditional alphas are too
volatile for one to reject the hypothesis that our chosen dynamic version of the Fama and
French (2015) factor model is a correct specification. The logic leads the econometrician
to infer that the average alpha spreads are consistent between Table 3 and Table A2.1.
4. Table A2.3 shows that each dynamic conditional HML beta exhibits much variability over
time for the original Fama-French value factor to be economically meaningful in explain-
ing time variation in average stock returns. In conjunction with the evidence of significant
long-term average HML betas in Table A5.3, the ARMA-GARCH results support the use of
HML as a relevant state variable that helps better span the investor’s mean-variance space.
Thus, HML conveys non-negligible information about at least some variation in average
returns for a wide variety of stock portfolio tilts. This major inference reconciles with some
recent independent empirical contributions of Fama and French (2015, 2016) and Hou et
al. (2015): HML appears to be redundant once the econometrician incorporates RMW and
CMA into the factor model for the U.S. stock market, whereas, the hefty value premium
persists both in the U.S. and several other stock markets (Asness et al. (2013); Fama and
French (2016)). In turn, the economic content and substance of HML and even SMB may

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 628 M.-W. Hung and A. J.-Y. Yeh

Table 4. ARMA-GARCH time-series representation of each dynamic conditional alpha and beta spread

Panel A ARMA(1,1)-EGARCH(1,1,1) ARMA(1,1)-GJR-GARCH(1,1,1)

Factor premium spread AR MA ARCH GARCH Expo AR MA ARCH GARCH GJR

Conditional alpha spread

.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Size          
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Momentum        
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Book-to-market       
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Cashflow-to-price     
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Dividend-to-price        
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Earnings-to-price        
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Investment       
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Profitability      
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Short-term reversal      
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Long-term reversal        
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Conditional MRP beta spread

.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Size      
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Momentum      
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Book-to-market        
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Cashflow-to-price      
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Dividend-to-price      
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Earnings-to-price     
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Investment      
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Profitability      
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Short-term reversal        
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Long-term reversal     

Panel B ARMA(1,1)-EGARCH(1,1,1) ARMA(1,1)-GJR-GARCH(1,1,1)

Factor premium spread AR MA ARCH GARCH Expo AR MA ARCH GARCH GJR

Conditional SMB beta spread

........................................................................................................................................................................................................................................................................

Size        
........................................................................................................................................................................................................................................................................

Momentum        
........................................................................................................................................................................................................................................................................

Book-to-market        
........................................................................................................................................................................................................................................................................

Cashflow-to-price      
........................................................................................................................................................................................................................................................................

Dividend-to-price      
........................................................................................................................................................................................................................................................................

Earnings-to-price     
........................................................................................................................................................................................................................................................................

Investment      
........................................................................................................................................................................................................................................................................

Profitability      
........................................................................................................................................................................................................................................................................

Short-term reversal        
........................................................................................................................................................................................................................................................................

Long-term reversal       
........................................................................................................................................................................................................................................................................

Conditional HML beta spread

........................................................................................................................................................................................................................................................................

Size      
........................................................................................................................................................................................................................................................................

Momentum        
........................................................................................................................................................................................................................................................................

Book-to-market      
........................................................................................................................................................................................................................................................................

Cashflow-to-price      
........................................................................................................................................................................................................................................................................

Dividend-to-price      
........................................................................................................................................................................................................................................................................

Earnings-to-price      

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 629

Table 4. Continued

Panel B ARMA(1,1)-EGARCH(1,1,1) ARMA(1,1)-GJR-GARCH(1,1,1)

Investment        
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Profitability    
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Short-term reversal      
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Long-term reversal     

Panel C ARMA(1,1)-EGARCH(1,1,1) ARMA(1,1)-GJR-GARCH(1,1,1)

Factor premium spread AR MA ARCH GARCH Expo AR MA ARCH GARCH GJR

Conditional RMW beta spread

........................................................................................................................................................................................................................................................................

Size        
........................................................................................................................................................................................................................................................................

Momentum     
........................................................................................................................................................................................................................................................................

Book-to-market         
........................................................................................................................................................................................................................................................................

Cashflow-to-price       
........................................................................................................................................................................................................................................................................

Dividend-to-price      
........................................................................................................................................................................................................................................................................

Earnings-to-price      
........................................................................................................................................................................................................................................................................

Investment      
........................................................................................................................................................................................................................................................................

Profitability         
........................................................................................................................................................................................................................................................................

Short-term reversal   
........................................................................................................................................................................................................................................................................

Long-term reversal     
........................................................................................................................................................................................................................................................................

Conditional CMA beta spread

........................................................................................................................................................................................................................................................................

Size        
........................................................................................................................................................................................................................................................................

Momentum      
........................................................................................................................................................................................................................................................................

Book-to-market        
........................................................................................................................................................................................................................................................................

Cashflow-to-price         
........................................................................................................................................................................................................................................................................

Dividend-to-price          
........................................................................................................................................................................................................................................................................

Earnings-to-price       
........................................................................................................................................................................................................................................................................

Investment       
........................................................................................................................................................................................................................................................................

Profitability        
........................................................................................................................................................................................................................................................................

Short-term reversal      
........................................................................................................................................................................................................................................................................

Long-term reversal     
This table summarizes the empirical results in Appendix 2 (cf. Tables A2.1 to A2.6). We demonstrate that each dynamic conditional factor premium
can be modeled as a typical financial time-series. We apply both ARMA(1,1)-EGARCH(1,1,1) and ARMA(1,1)-GJR-GARCH(1,1,1) conditional-mean-and-
variance models to fit each factor premium that the econometrician extracts from a “dynamic” variant of the Fama and French (2015) factor model.
Although it is possible to identify a more precise time-series representation for each factor premium, our goal here is more straight-forward. In fact,
our primary and ultimate goal is to use the standard toolkit in time-series econometrics to establish the empirical fact that each dynamic conditional
alpha or beta spread exhibits the major properties of most financial time series. Each factor premium embeds autoregressive mean reversion in the
conditional mean specification of ARMA(1,1), as well as volatility clusters and asymmetries in the conditional volatility specification of EGARCH(1,1,1)
or GJR-GARCH(1,1,1) (cf. Engle (1982); Bollerslev (1986); Nelson (1991); Glosten et al (1993)).

help explain whether these state variables serve as useful empirical proxies for macroeco-
nomic innovations (Liew and Vassalou (2000), Vassalou (2003); Petkova (2006); Hahn and
Lee (2006)), distress risk (Griffin and Lemmon (2002); Vassalou and Xing (2004)), or some
other behavioral mispricing reasons (Campbell et al. (2008)). Our subsequent evidence
generalizes the key empirical inference that mutual causation between macroeconomic
surprises and dynamic conditional alpha spreads can be an informative and plausible eco-
nomic explanation for most anomalies in the fundamental evolution of average returns
and factors.

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 630 M.-W. Hung and A. J.-Y. Yeh

3.4 Conditional specification test evidence

In this section, we derive and design a new conditional specification test to differentiate the
static and dynamic conditional models. Under the null hypothesis, both the static and dynamic
conditional estimators are consistent and asymptotically normal while the former attains the
Cramer-Rao lower bound and therefore is efficient in the standard econometric nomenclature
with the parameter vector θ={α, β m , β s , β h , β r , β c }. In contrast, only the dynamic conditional
estimator is consistent under the alternative hypothesis. For our subsequent derivation, we define
the distance between the static and dynamic conditional estimators as ξ = θ d − θ s . The null
hypothesis suggests a trivial distance between the static and dynamic conditional estimators:
ξ̂ = θ̂ d − θ̂ s ⇒ H0 : θ d = θ s ⇒ H0 : ξ = 0 (18)
We define the variance-covariance matrix for the main distance between each estimator and
its respective true parameter vector V s and V d where equations (1) and (2) carry the Gaussian
normal error terms ε t and et :
 
N 1/2 · θ̂ s − θ s ∼ N(0, V s ) (19)
 
N 1/2 · θ̂ d − θ d ∼ N(0, V d ) (20)
The proof begins with the naïve assumption the static estimator is not orthogonal to the dis-
tance between the static and dynamic conditional estimators. Then, the mathematical derivation
leads to a contradiction to affirm the alternative scenario that the static estimator is indeed orthog-
onal to the vector wedge between the static and dynamic conditional estimators. The key crux of
this simple proof relies on the Cramer-Rao lower-bound efficiency criterion that the vector dis-
tance between the static and dynamic variance-covariance matrices must be positive-semidefinite.
In fact, we can use a unique algorithm to handle the rare case where the vector distance is
negative-semidefinite. We first define the dynamic conditional estimator as the static estimator
plus some multiple of the above vector wedge in equation (18) while the latter term must be zero
under the null hypothesis (where we expect the scalar k = 0 with the probability limit plim ξ = 0
under the null hypothesis that both static and dynamic conditional estimators are asymptotically
equivalent):
θ̂ d = θ̂ s + kMξ̂ (21)
where k is a scalar and M is a chosen matrix. Next, we expand the variance-covariance matrix for
the dynamic conditional estimator:
       
V θ̂ d = V θ̂ s + kMQ + kMT QT + k2 MV ξ̂ MT ⇒ Q = cov θ̂ s , ξ̂ (22)
We can also expand the vector wedge between the respective variance-covariance matrices for
the static and dynamic conditional estimators:
       
W k = V θ̂ d − V θ̂ s = kMQ + kQT MT + k2 MV ξ̂ MT (23)

We derive the first-order and second-order derivatives of this vector wedge with respect to k:
   
∇k W k = MQ + QT MT + 2kMV ξ̂ MT (24)
   
∇k2 W k = 2MV ξ̂ MT (25)
Equation (25) suggests that the second-order derivative is positive definite. As a result, one can
figure out the minimum vector wedge when he or she sets the first-order condition (24) to zero:
      −1
∇k W k = MQ + QT MT + 2kMV ξ̂ MT = 0 ⇒ M = − 2kV ξ̂ QT (26)

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 631

By definition, we know the equality W(k = 0)=0. When we evaluate the first-order derivative
at k = 0, this resultant matrix cannot be positive-semidefinite:
    −1   −1
∇k W k = 0 = MQ + QT MT = − 2kV ξ̂ 2QT Q = − kV ξ̂ QT Q (27)
Equation (27) suggests that the resultant matrix must not be positive-semidefinite. This solu-
tion poses a contradiction to the assumption that W(k) must be positive-semidefinite. For
arbitrarily small values of k, equation (27) suggests that W(k) can be negative-semidefinite.
In turn, this contradiction suggests that Q equates a null matrix. Therefore, we can find zero
covariance between the static estimator and its distance from the dynamic conditional estimator.
This analytical result allows us to derive a new dynamic conditional specification test in a
similar spirt of the econometric contributions of Durbin (1954), Wu (1973), Hausman (1978),
Nakamura and Nakamura (1981), White (1982), Newey (1985), and Tauchen (1985).
           
ξ̂ = θ̂ d − θ̂ s ⇒ ξ̂ + θ̂ s = θ̂ d ⇒ V ξ̂ + V θ̂ s = V θ̂ d ⇒ V ξ̂ = V θ̂ d − V θ̂ s (28)

T
  −1 T
    −1  
κ = ξ̂ T −1 V̂ ξ̂ ξ̂ = T · ξ̂ V θ̂ d − V θ̂ s ξ̂ ∼ χ 2 q (29)

where q denotes the number of parameters as the degrees of freedom for the χ 2 test statistic and
T denotes the number of time periods over the entire data span.
While this dynamic conditional specification test statistic shares the same sandwich form of
most econometric tests, the middle matrix may or may not always be positive-semidefinite. This
occurrence arises from the rare but plausible case where the static OLS estimator contains much
measurement noise that the dynamic conditional estimator becomes similar in terms of efficiency.
In this rare case, we need to adjust this unconventional middle matrix to the nearest positive def-
inite matrix (Higham (2002)). This algorithmic approach entails setting any negative eigenvalues
to zero to re-approximate the positive-definite matrix. Although we sometimes run this algorithm
to carry out the dynamic conditional specification test, this extreme but plausible scenario is rather
rare. When this scenario occurs in practice, the resultant dynamic conditional specification test
statistics may not exceed the critical value and so there is insufficient evidence to differentiate
the static and dynamic estimators. In practice, our conditional specification test rejects the null
hypothesis of a correct static model for at least 95% of all the stock decile tilts. This rejection sug-
gests that the alternative dynamic conditional alpha model yields consistent estimates of factor
premiums.
Table 5 lists the conditional specification test evidence in favor of the dynamic conditional
multifactor asset pricing model. In our empirical analysis of 100 monthly decile returns on the
major anomalies plus their respective 10 long-short portfolios that focus on the extreme deciles,
95% of the portfolio tilts suggest the statistical rejection of the null hypothesis that the static
factor model is a correctly specification. The preponderance of empirical results supports the
dynamic conditional factor model in stark contrast to the static baseline model for good reasons.
First, the vast majority of dynamic conditional alphas are statistically insignificant so that most
of the pricing errors are close to zero in the alternative dynamic conditional model specification.
Second, the conditional specification test results prevail in favor of the alternative hypothesis
that only the dynamic conditional estimator is consistent. This latter point highlights the power
of measurement error minimization that the recursive multivariate filter achieves for consistent
statistical estimation. Overall, the dynamic conditional factor model can outperform its static
counterpart in light of the conditional specification test evidence and dynamic conditional alpha
insignificance.4
Therefore, both the recursive multivariate filter and dynamic conditional specification test
add value to the econometric toolkit for subsequent asset-pricing analysis. Not only does this
econometric advancement pose a core conceptual challenge to the conventional use of ordinary

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 632 M.-W. Hung and A. J.-Y. Yeh

Table 5. Conditional specification test of the static versus dynamic conditional multifactor asset pricing models

Portfolio Low Decile 2 Decile 3 Decile 4 Decile 5 Decile 6 Decile 7 Decile 8 Decile 9 High Spread
Size
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

κ test statistic 32.0 178.8 42.2 46.8 257.5 4092.7 166.9 164.4 13,200.4 0.0 31.8
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.999 0.001
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Momentum
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

κ test statistic 33.0 42.0 41.0 13.2 829.6 0.0 20.4 11,327.4 49.5 61.4 3.1
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.001 0.001 0.001 0.067 0.001 0.999 0.005 0.001 0.001 0.001 0.876
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Book-to-market
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

κ test statistic 18.5 47.3 191.6 247.1 160.6 1371.7 710.3 477.8 580.3 462.0 14.6
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.0097 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.042
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Cashflow-to-price
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

κ test statistic 24.2 11.9 324.3 220.1 345.3 189.0 214.6 553.9 1111.8 407.2 223.0
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.001 0.104 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Dividend-to-price
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

κ test statistic 59.0 108.3 102.3 53.2 440.0 1623.9 262.1 1655.0 512.9 254.7 222.5
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Earnings-to-price
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

κ test statistic 10.3 26.6 169.0 821.0 149.8 200.3 385.0 495.1 748.0 339.8 87.0
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.171 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Investment
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

κ test statistic 15.6 206.4 33.5 2417.7 867.0 1017.3 1452.1 683.6 1102.5 521.7 23.9
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.029 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Profitability
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

κ test statistic 20.5 398.8 209.1 162.2 203.8 4215.2 743.9 830.9 1938.1 4744.2 23.0
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.005 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.002
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Short-term reversal
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

κ test statistic 28.4 744.3 168.8 93.2 76.0 2.3 1331.4 60.8 1047.9 319.1 221.4
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.001 0.001 0.001 0.001 0.001 0.940 0.001 0.001 0.001 0.001 0.001
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Long-term reversal
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

κ test statistic 20.4 156.6 455.9 899.0 10,916.2 598.9 344.1 844.1 4923.1 2151.9 27.5
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

p-value 0.005 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
Over the 50-year period from January 1964 to December 2013, one applies the recursive multivariate Filter to extract dynamic multifactor factor
premiums from the Fama and French (2015) five-factor asset pricing model. At each time increment, the econometrician takes into account the
Fama and French (2015) factors such as the excess return on the market portfolio (MRP), the return spread between the top 30% small and bottom
30% big stocks (SMB), the return spread between the top 30% high book-to-market and bottom 30% low book-to-market stocks (HML), the return
spread between the top 30% robust and bottom 30% weak stocks in terms of their relative profitability (RMW), and the return spread between
the top 30% conservative investment and bottom 30% aggressive investment stocks (CMA) to explain the variation in the excess return on each
stock decile for size, momentum, value (cf. book-to-market, cashflow-to-price, dividend-to-price, and earnings-to-price), investment, profitability,
short-term return reversal, and long-term return reversal. The econometrician presents the mathematical time-series representation below:.

Rkt − Rft = αt + βmt (Rmt − Rft ) + βst SMBt + βht HMLt + βrt RMWt + βct CMAt + εt

Table 5 presents the conditional specification test evidence for each stock decile sorted on size, value, momentum, asset investment, operating
profitability, short-term return reversal, and long-term return reversal. The first 10 columns summarize the χ 2 -statistic for each stock decile and its
corresponding p-value for the null hypothesis of a consistent and efficient static estimator against the alternative hypothesis of a correct consistent
dynamic conditional estimator. The next column shows the χ 2 -statistic for each long-short top-bottom stock decile and its corresponding p-value.
About 95% of the χ 2 -statistics are econometrically significant with at least 90%+ confidence. The appendix depicts the dynamic conditional alpha
time-series for each of the stock portfolio tilts that yield anomalous excess returns in static asset pricing analysis (i.e. size, value, momentum, asset
investment, operating profitability, short-term return reversal, and long-term return reversal).

ξ̂ = θ̂ d − θ̂ s ⇒ H0 : θ d = θ s ⇒ H0 : ξ = 0

1/2
· (θ̂ s − θ s ) ∼ N(0, V s ) N1/2 · (θ̂ d − θ d ) ∼ N(0, V d )
N
           
ξ̂ = θ̂ d − θ̂ s ⇒ ξ̂ + θ̂ s = θ̂ d ⇒ V ξ̂ + V θ̂ s = V θ̂ d ⇒ V ξ̂ = V θ̂ d − V θ̂ s

T
  −1 T
    −1
κ = ξ̂ T −1 V̂ ξ̂ ξ̂ = T · ξ̂ V θ̂ d − V θ̂ s ξ̂ ∼ χ 2 (q)

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 633

least-squares (OLS) regressions for factor model design, but this econometric innovation also
suggests that dynamic conditional alpha estimation serves as a useful quantitative method for
financial applications such as mutual fund performance evaluation, corporate event assessment,
equity cost estimation, and financial risk management.

4. Mutual causation between macroeconomic innovations and alpha spreads

In this section, we delve into the main crux of the current study and empirically ascertain mutual
causation between macroeconomic innovations and dynamic conditional alpha spreads. To the
extent that macroeconomic innovations manifest in the form of these dynamic conditional alpha
spreads, this causality reveals the marginal investor’s fundamental news and expectations about
the cross-section of average returns. We interpret this Granger-causality evidence in the broader
context of the intertemporal CAPM with cash-flow news, discount-rate news, as well as future-
risk news (cf. Merton (1973); Campbell (1993); Campbell and Vuolteenaho (2004); Campbell et
al. (2017)). Campbell and Vuolteenaho (2004) develop the intertemporal CAPM with cash-flow
and discount-rate betas. Campbell et al. (2017) extend and generalize the intertemporal CAPM
with stochastic volatility to capture future-risk news. The price of risk for cash-flow news is the
marginal investor’s relative risk aversion coefficient times more than the unit price of risk for
negative discount-rate news. Therefore, cash-flow news carries “bad betas”, whereas, negative
discount-rate news carries “good betas”. Also, an asset that provides positive returns when future
risk expectations increase tends to generate low average returns. Thus, the marginal investor’s
stochastic discount factor is high when he or she anticipates high stochastic volatility in the future.
In essence, these fundamental news and expectations reflect the marginal investor’s rational
response to different kinds of macroeconomic surprises with respect to cash flows, discount rates,
and stochastic volatilities. Vector autoregressions (VAR) and Granger-causality tests below accord
with the main theme of bilateral causation between macroeconomic innovations and dynamic
conditional alphas. This causation reinforces the intertemporal asset- pricing interpretation that
macroeconomic shocks manifest in the form of dynamic conditional factor premiums and vice
versa such that this nexus reveals the marginal investor’s fundamental news and macroeconomic
expectations about the cross-section of average returns. To the extent that these causal relations
reflect the marginal investor’s rational response to changes in his or her intertemporal choice
and conditional expectation of terminal wealth, the resultant dynamic conditional factor model
differentiates itself from most behavioral mispricing models. This key conceptual distinction res-
onates with the primary thesis of Kozak et al. (2018) recent critique of numerous horseraces for
empirically-driven factor models.
We use 15 main monthly time-series in a macroeconometric vector autoregression (VAR)
(Sims (1980); Campbell (1993)). There are 12 macro time-series, 2 financial uncertainty met-
rics, and 2 investor sentiment proxies. The resultant dataset spans the 285-month sample period
from April 1990 to December 2013. These macrofinancial time-series include first differences in
the national economic activity index, Treasury bill rate, unemployment rate, term spread, default
spread, prime bank loan rate, aggregate equity market dividend yield, as well as percent changes
in industrial production, non-farm payroll, house price index, consumer price index, exchange
rate, financial stress index, economic policy uncertainty, and investor sentiment. For the Baker-
Wurgler capital market investor sentiment index, we use the first principal component as a main
empirical proxy. This variable choice has no impact on our subsequent inferences. Table 6 lists
and describes these macroeconomic variable definitions and their data sources.
We develop a medium macroeconometric vector autoregressive system in order to gauge
macroeconomic innovations or fundamental surprises that the typical investor would face in his
or her investment journey. Insofar as we can gauge macroeconomic surprises, we establish the
empirical fact of mutual Granger causation between macroeconomic innovations and dynamic

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 634 M.-W. Hung and A. J.-Y. Yeh

Table 6. Macroeconomic variable definitions and their data sources

Macroeconomic variable and name definition Source

Chicago Fed’s national economic activity index change from historical trend Chicago Fed
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

St. Louis Fed Treasury 3-month secondary-market bill rate change St. Louis Fed
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

St. Louis Fed unemployment rate change (total unemployment/labor force participation) St. Louis Fed
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Term spread between the 10-year Treasury and 3-month Treasury constant maturity rates St. Louis Fed
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Default spread between Moody’s Baa-corporate-bond and 10-year Treasury bill rates St. Louis Fed
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Prime bank loan rate change for Top 25 U.S. commercial banks in terms of total assets St. Louis Fed
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

S&P 500 dividend yield from Professor Robert Shiller’s book on irrational exuberance Shiller
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

St. Louis Fed national industrial production index change with the base year in 2007 St. Louis Fed
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Bureau of Labor Statistics non-farm payroll (in thousands of persons) percent change Bureau of Labor Statistics
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Freddie Mac U.S. metropolitan-area residential house price index percent change Freddie Mac
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Bureau of Labor Statistics consumer price index (for urban consumers) percent change Bureau of Labor Statistics
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Federal Reserve U.S. trade-weighted average composite dollar index percent change Federal Reserve Board
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Baker-Bloom-Davis U.S. economic policy uncertainty index percent change Baker et al. (2016)
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Kansas City Fed financial stress index change from 11 key financial market variables Kansas City Fed
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Baker-Wurgler investor sentiment percent change (from the first principal component) Baker and Wurgler (2007)
This table describes the macroeconomic variable definitions and their public data sources in our vector autoregression analysis of Granger mutual
causation between fundamental macro surprises and dynamic conditional factor premiums. To the extent that macro surprises manifest in the
form of dynamic conditional factor premiums, this Granger causation reveals the marginal investor’s fundamental news and rational expectations
about the cross-section of average returns.
We specify 15 major monthly time-series in a standard macroeconometric vector autoregressive system. There are 12 macroeconomic vari-
ables, 2 financial uncertainty time-series, and 2 investor sentiment proxies. For the Baker-Wurgler investor sentiment index, we use the original
first principal component as a better empirical proxy. The resultant dataset spans the 285-month sample period from April 1990 to December
2013. These macroeconomic time-series include changes in the national economic activity index, Treasury bill rate, unemployment rate, term
spread, default spread, prime bank loan rate, aggregate equity market dividend yield, and percent changes in U.S. industrial production, non-farm
payroll, house price index, consumer price index, exchange rate, financial stress index, economic policy uncertainty, and investor sentiment.

conditional alpha spreads. Granger-causality tests help us assess this mutual causation as a core
qualifying condition for fundamental factor selection.
Table 7 shows the vector autoregression (VAR) coefficient estimates and t-statistics. This VAR
model explains most of the time variation in fundamental macroeconomic news. The only excep-
tion is the macro VAR model of percent changes in the economic policy uncertainty index (Baker
et al. (2016)), which seems to contain peculiar information in its own right. Due to its unique lag
structure, the Baker and Wurgler (2007) financial market investor sentiment VAR model gener-
ates the highest R2 of 94.9%. This macroeconometric VAR system captures reasonably well time
variation in U.S. national economic activity, prime bank loan interest rate, non-farm payroll, and
residential house price index with intermediate R2 s from 40% to 68%.
This macroeconometric VAR system explains about 20–30% of time variation in each of the
other key macroeconomic time-series such as the Treasury 3-month bill rate, unemployment rate,
term spread, default spread, aggregate dividend yield, industrial production, dollar strength, CPI
inflation, and systemic stress within the U.S. financial sector. Overall, this medium macro VAR
system allows us to extract macroeconomic innovations (or fundamental surprises) as the residual
disturbances from each of the resultant macro equations.
Tables 8 and 9 list the pairwise Granger causality tests with p-values for us to see whether
there is mutual causation between macroeconomic surprises and dynamic conditional alphas.
Indeed, these test results suggest the affirmative case for Granger causation in both directions.
Table 8 shows that at least 1 to 4 macroeconomic innovations Granger-cause each dynamic con-
ditional alpha spread for size, value (B/M, CF/P, and Div/P), operating profitability, asset growth,
short-term return reversal, and long-run return reversal. Only momentum and partial value (E/P)
turn out to be the exceptions that defy this rule of thumb.

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

Table 7. Vector AutoRegression (VAR) of macroeconomic fluctuations with consistent coefficient estimates and t-statistics

Macroeconomic variable (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)

Intercept −45.809 13.923 −52.542 −3.716 7.332 12.378 −0.037 0.029 −0.376 0.509 −0.565 −0.094 −1.009 29.052 23.922
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

National activity index change −0.447 −0.002 0.033 0.027 0.003 −0.008 0.018 −0.002 0.000 0.001 −0.001 0.002 0.006 0.011 −0.013
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Treasury 3-month bill rate change 0.218 0.210 0.032 −0.251 −0.057 0.450 0.098 0.002 0.001 0.001 0.000 −0.001 −0.068 −0.045 −0.014
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Unemployment rate change 0.034 −0.035 −0.152 0.135 −0.113 −0.035 −0.048 −0.005 −0.002 0.000 0.001 −0.003 −0.022 −0.262 −0.012
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Term spread (10-year minus 1-year) 0.269 −0.046 −0.065 0.123 −0.121 0.006 −0.004 0.000 0.000 0.002 −0.001 −0.015 −0.075 −0.331 0.052
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Default spread (Baa minus 10-year) −0.589 −0.207 0.080 0.036 0.155 −0.103 −0.108 −0.009 0.000 0.001 −0.001 −0.023 −0.005 0.007 −0.052
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Prime bank loan rate change 0.261 0.117 −0.106 0.065 0.014 0.203 −0.069 0.002 0.000 0.000 0.000 −0.014 0.087 0.006 0.007
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Aggregate S&P 500 dividend yield −1.318 −0.119 −0.041 −0.415 0.617 −0.046 0.255 −0.014 −0.002 0.000 −0.001 −0.013 0.118 0.396 0.101
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Industrial production index change −27.389 2.535 −3.206 −5.961 −5.808 2.264 −4.374 0.128 0.045 −0.080 0.052 −0.600 −0.313 −20.078 −1.524
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Non-farm payroll percent change −39.774 11.453 −47.308 −2.925 10.034 9.044 2.895 0.691 0.554 0.184 0.100 1.065 −4.478 35.224 16.951
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

House price index percent change 5.901 1.205 −1.330 −1.440 −1.066 1.067 0.408 0.051 0.006 0.870 −0.018 0.002 1.846 −0.471 1.211
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Consumer price index percent change 14.084 −0.937 −0.426 6.208 3.044 0.443 0.988 0.158 0.022 0.509 0.333 0.051 3.344 12.995 6.599
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Dollar trade index percent change 1.104 −0.347 −0.400 0.519 1.117 −0.397 0.034 −0.003 −0.003 0.028 −0.033 0.392 −0.172 1.299 0.704
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Economic uncertainty percent change 0.214 0.037 0.081 −0.111 0.014 −0.027 0.012 0.003 0.000 −0.001 0.000 −0.004 −0.218 0.107 0.001
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Financial stress index change 0.219 0.056 0.009 0.075 0.073 0.030 0.102 0.005 0.000 0.001 −0.002 0.017 −0.010 0.048 −0.060
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Baker-Wurgler investor sentiment −0.007 −0.042 0.022 0.015 0.021 −0.023 0.009 0.000 0.000 0.000 0.000 0.001 0.008 0.025 0.963

Macroeconomic variable (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)

Macroeconomic Dynamics
Intercept −1.68 1.67 −7.41 −0.54 0.85 2.00 −0.19 0.07 −7.18 2.31 −4.68 −0.16 −0.14 1.69 2.58
.............................................................................................................................................................. ............................................................................................................................................................................................................................

National activity index change −7.97 −0.11 2.35 1.55 0.17 −0.62 2.22 −3.08 −3.69 1.68 −2.70 2.06 0.34 0.36 −0.43
.............................................................................................................................................................. ............................................................................................................................................................................................................................

Treasury 3-month bill rate change 0.80 2.41 0.41 −2.82 −0.71 7.20 2.44 0.61 1.61 0.53 0.25 −0.14 −0.81 −0.26 −0.57
.............................................................................................................................................................. ............................................................................................................................................................................................................................

Unemployment rate change 0.14 −0.53 −2.53 1.83 −1.62 −0.71 −1.37 −1.94 −3.65 0.21 1.12 −0.65 −0.30 −1.92 −0.08
.............................................................................................................................................................. ............................................................................................................................................................................................................................

Term spread (10-year minus 1-year) 1.21 −0.65 −1.19 1.75 −1.87 0.14 −0.11 −0.05 −0.50 1.24 −0.87 −3.33 −1.12 −2.57 −0.09
.............................................................................................................................................................. ............................................................................................................................................................................................................................

Default spread (Baa minus 10-year) −1.99 −2.25 1.10 0.39 1.82 −1.58 −2.49 −2.91 0.28 0.25 −0.65 −3.94 −0.05 0.04 −1.11
.............................................................................................................................................................. ............................................................................................................................................................................................................................

Prime bank loan rate change 1.02 1.46 −1.71 0.80 0.15 3.49 −1.76 0.58 0.39 0.21 −0.22 −2.72 1.09 0.05 0.32
.............................................................................................................................................................. ............................................................................................................................................................................................................................

Aggregate S&P 500 dividend yield −2.62 −0.72 −0.32 −2.64 4.21 −0.38 3.41 −2.63 −2.25 0.09 −0.59 −1.33 0.77 1.34 0.71

635
https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

636
M.-W. Hung and A. J.-Y. Yeh
Table 7. Continued

Macroeconomic variable (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)

Industrial production index change −3.93 1.14 −1.86 −2.70 −2.88 1.44 −4.27 1.73 3.46 −1.48 1.75 −4.38 −0.14 −5.01 −0.89
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Non-farm payroll percent change −1.49 1.41 −6.90 −0.42 1.23 1.51 0.60 2.34 10.57 0.87 0.90 1.98 −0.56 2.17 2.02
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

House price index percent change 1.14 0.78 −1.01 −0.95 −0.74 0.94 0.49 0.94 0.59 22.10 −0.79 0.04 1.21 −0.20 0.82
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Consumer price index percent change 0.91 −0.14 −0.09 1.26 0.68 0.16 0.36 0.98 0.69 4.36 5.30 0.17 0.73 1.46 1.25
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Dollar trade index percent change 0.35 −0.45 −0.46 0.56 1.39 −0.68 0.04 −0.13 −0.57 1.27 −2.69 7.04 −0.19 0.78 0.28
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Economic uncertainty percent change 0.99 0.53 1.51 −1.63 0.24 −0.58 0.40 1.20 0.14 −0.89 0.45 −0.88 −3.36 0.87 −0.41
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Financial stress index change 1.44 1.19 0.20 1.56 1.63 0.91 4.59 3.27 0.11 0.86 −3.16 5.63 −0.22 0.54 −1.82
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Baker-Wurgler investor sentiment 0.06 −2.50 1.29 1.14 1.31 −1.86 1.48 −0.22 1.64 −0.34 0.58 1.10 0.50 0.99 54.63
This table summarizes the macroeconometric vector autoregression (VAR) results. The upper panel presents the consistent coefficient estimates, and the lower panel shows the corresponding t-statistics. We
use 15 main monthly time-series in a macroeconometric vector autoregressive system (Sims (1980); Campbell (1993)). There are 12 macro time-series, 2 financial uncertainty metrics, and 2 investor sentiment
proxies. The resultant dataset spans the 285-month sample period from April 1990 to December 2013. These macro time-series encompass first differences in the national economic activity index, Treasury bill
rate, unemployment rate, term spread, default spread, prime bank loan rate, aggregate dividend yield, and percent changes in industrial production, non-farm payroll, house price index, consumer price index,
exchange rate, financial stress index, economic policy uncertainty, and investor sentiment. Appendix 3 presents and describes the macroeconomic variable definitions and their data sources.
 Macroeconomic Dynamics 637

Table 8. Granger causation from macroeconomic innovations to dynamic conditional alpha spreads

Macroeconomic variable Size Momen B/M CF/P Div/P E/P AG OP SR Rev LR Rev

National activity index change 0.65 0.13 0.43 0.40 0.067 0.18 0.21 0.73 0.82 0.70
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Treasury 3-month bill rate change 0.105 0.90 0.54 0.72 0.15 0.94 0.021 0.98 0.38 0.026
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Unemployment rate change 0.64 0.62 0.33 0.075 0.35 0.79 0.034 0.56 0.13 0.98
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Term spread (10-year minus 1-year) 0.78 0.96 0.29 0.61 0.63 0.64 0.12 0.44 0.41 0.0317
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Default spread (Baa minus 10-year) 0.16 0.52 0.32 0.15 0.14 0.18 0.027 0.64 0.88 0.0321
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Prime bank loan rate change 0.19 0.43 0.59 0.95 0.46 0.86 0.21 0.86 0.16 0.84
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Aggregate S&P 500 dividend yield 0.25 0.56 0.52 0.32 0.19 0.21 0.49 0.81 0.072 0.54
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Industrial production index change 0.52 0.88 0.72 0.98 0.58 0.27 0.78 0.35 0.92 0.16
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Non-farm payroll percent change 0.098 0.69 0.86 0.35 0.99 0.31 0.40 0.48 0.11 0.64
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

House price index percent change 0.23 0.14 0.14 0.47 0.78 0.99 0.78 0.26 0.40 0.25
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Consumer price index percent change 0.029 0.63 0.073 0.15 0.14 0.55 0.86 0.63 0.23 0.93
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Dollar trade index percent change 0.60 0.88 0.45 0.98 0.11 0.69 0.35 0.35 0.70 0.51
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Economic uncertainty percent change 0.98 0.79 0.21 0.66 0.62 0.61 0.33 0.61 0.084 0.47
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Financial stress index change 0.86 0.16 0.32 0.1002 0.013 0.44 0.037 0.57 0.65 0.005
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Baker-Wurgler investor sentiment 0.73 0.57 0.67 0.27 0.68 0.40 0.12 0.056 0.76 0.55
This table shows the pairwise Granger-causality test p-values for the econometrician to assess whether Granger causation runs from macroe-
conometric innovations to dynamic conditional alphas. The columns indicate the respective asset pricing anomalies such as size, momentum,
value (B/M, CF/P, Div/P, and E/P), asset growth, operating profitability, short-run return reversal, and long-run return reversal. This tables demon-
strates that at least a few macro innovations Granger-cause each dynamic conditional alpha spread with at least 90% statistical confidence
(p-values < 0.10) for most anomalies such as size, value (B/M, CF/P, and Div/P), asset growth, operating profitability, short-run return reversal,
and long-term reversal. Only momentum and partial value (E/P) turn out to be the exceptions that defy this rule of thumb. Within the intertempo-
ral CAPM context, we interpret this evidence in support of the prevalent use of Fama and French (2015) factors in a dynamic conditional model.
Granger causation between macroeconomic innovations and dynamic conditional alphas serves as a core qualifying condition for “shrinking the
factor zoo”, whereas, there is minimal evidence in favor of viewing momentum as a conceptually sound fundamental factor in the dynamic condi-
tional context. To the extent that the investor cannot decipher fundamental news nor macroeconomic surprises from momentum, it remains hard
to rationalize momentum apart from a unique statistical aberration. Insofar as momentum returns persist as an anomaly, we need a more plausi-
ble explanation than behavioral mispricing disequilibrium before we view momentum as an extra fundamental factor in the dynamic conditional
model.

Table 9 shows that reverse causality runs from several dynamic conditional alpha spreads to
macroeconomic innovations. Key conditional alpha spreads represent rich and valuable stock
market signals about macroeconomic surprises. However, this evidence is less conclusive since
only partial value (Div/P and E/P), asset investment growth, operating profitability, short-term
return reversal, and long-term return reversal exhibit this reverse causation. In fact, the primary
conditional alpha spreads for size, value (B/M and CF/P), and momentum convey little causal
information about macroeconomic innovations.
We retain our main focus on the recursive multivariate filtration of Fama-French dynamic
conditional alphas. The subsequent GARCH and VAR results complement and corroborate this
focus. For this reason, we would like to draw attention to mutual causation between conditional
alphas and macro innovations in Table 9.5 Specifically, many stock market alphas help predict
macroeconomic surprises: partial value alpha spreads Div/P and E/P show this critical causation.
This evidence resonates with the title of the current study. We highlight this reverse causation
in Table 9 while the vector autoregression results in Table 8 support the conventional view that
macroeconomic surprises Granger-cause at least some stock market alphas.
On balance, these results suggest mutual Granger causation from macroeconomic surprises to
most dynamic conditional alpha spreads and vice versa (except momentum and partial value).
Within the intertemporal CAPM context, we interpret this evidence in support of the prevalent
use of Fama and French (2015) factors in a dynamic conditional model. Mutual causation between
macroeconomic innovations and dynamic conditional alpha spreads serves as a core qualifying
condition for “shrinking the factor zoo”, whereas, there is minimal or no sufficient evidence in

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 638 M.-W. Hung and A. J.-Y. Yeh

Table 9. Granger causation from dynamic conditional alpha spreads to macroeconomic innovations

Macroeconomic variable Size Momen B/M CF/P Div/P E/P AG OP SR Rev LR Rev

National activity index change 0.32 0.52 0.31 0.20 0.64 0.087 0.80 0.66 0.91 0.28 ∗
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Treasury 3-month bill rate change 0.87 0.34 0.46 0.96 0.016 0.72 0.47 0.63 0.92 0.44 ∗
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Unemployment rate change 0.22 0.35 0.27 0.28 0.008 0.017 0.49 0.44 0.98 0.70 ∗
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Term spread (10-year minus 1-year) 0.16 0.24 0.28 0.71 0.57 0.20 0.27 0.22 0.43 0.54
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Default spread (Baa minus 10-year) 0.70 0.80 0.73 0.81 0.46 0.88 0.37 0.98 0.35 0.83
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Prime bank loan rate change 0.22 0.20 0.68 0.83 0.31 0.34 0.50 0.0963 0.82 0.0916 ∗
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Aggregate S&P 500 dividend yield 0.15 0.80 0.64 0.72 0.61 0.77 0.74 0.87 0.62 0.86
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Industrial production index change 0.14 0.16 0.31 0.11 0.95 0.06 0.86 0.82 0.87 0.37 ∗
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Non-farm payroll percent change 0.93 0.57 0.42 0.11 0.04 0.08 0.58 0.22 0.28 0.54 ∗
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

House price index percent change 0.41 0.20 0.52 0.34 0.29 0.84 0.75 0.64 0.12 0.39
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Consumer price index percent change 0.50 0.30 0.46 0.97 0.90 0.73 0.51 0.81 0.022 0.98 ∗
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Dollar trade index percent change 0.61 0.52 0.70 0.71 0.54 0.67 0.82 0.43 0.30 0.96
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Economic uncertainty percent change 0.54 0.30 0.99 0.23 0.52 0.43 0.18 0.32 0.97 0.45
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Financial stress index change 0.20 0.91 0.41 0.92 0.61 0.60 0.96 0.52 0.26 0.56
.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Baker-Wurgler investor sentiment 0.11 0.52 0.33 0.71 0.97 0.63 0.035 0.029 0.001 0.98 ∗
This table shows the pairwise Granger-causality test p-values for the econometrician to assess whether Granger causation runs from dynamic
conditional alphas to macroeconometric innovations. The columns indicate the respective asset pricing anomalies such as size, momentum, value
(B/M, CF/P, Div/P, and E/P), asset growth, operating profitability, short-run return reversal, and long-run return reversal. Only partial value (Div/P
and E/P), asset growth, operating profitability, short-run return reversal, and long-run reversal exhibit this reverse causation (p-values < 0.10). In
fact, the primary conditional alpha spreads for size, value (B/M and CF/P), and momentum convey little causal information about macroeconomic
innovations (p-values > 0.10). Within the intertemporal CAPM context, we interpret this evidence in support of the prevalent use of Fama and French
(2015) fundamental factors in a dynamic conditional model. Granger causation between macroeconomic innovations and dynamic conditional
alphas serves as a core qualifying condition for “shrinking the factor zoo”, whereas, there is minimal evidence in favor of viewing momentum
as a conceptually sound fundamental factor in the dynamic conditional context. To the extent that the investor cannot decipher fundamental
news nor macroeconomic surprises from momentum, it remains hard to rationalize momentum apart from a unique statistical aberration. Insofar
as momentum profits persist as an anomaly, we need a more plausible explanation than behavioral mispricing disequilibrium before we view
momentum as an extra fundamental factor in the dynamic conditional model.

favor of viewing momentum as a conceptually sound factor in the dynamic conditional context.
To the extent that the marginal investor cannot decipher fundamental news nor macroeconomic
surprises from momentum, it is difficult to rationalize momentum apart from a unique statistical
aberration. Insofar as momentum profits persist as an anomaly, we need a more plausible reason
than behavioral mispricing disequilibrium before we view momentum as an extra fundamental
factor in the dynamic conditional model.
Tables 8 and 9 demonstrate bilateral causation between most macroeconomic shocks and
dynamic conditional alpha spreads. To the extent that macroeconomic innovations manifest in
the form of these dynamic conditional alpha spreads, this critical channel of Granger causation
reveals the marginal investor’s fundamental news and expectations about the cross-section of
average returns. This core evidence enriches and contributes to our chosen intertemporal asset-
pricing interpretation of dynamic conditional factor models. In this intertemporal asset-pricing
context, macroeconomic innovations serve as fundamental news and surprises that induce cash-
flow betas and future-risk betas as “bad betas” or negative discount-rate betas as “good betas”.
In accordance with this intertemporal CAPM thesis, we would expect assets with positive cash-
flow shocks, future-risk spillovers, or subpar discount-rate news to generate low average returns.
Conversely, we would expect other assets with negative cash-flow surprises, volatility declines,
or optimistic discount-rate news to generate high average returns. Therefore, mutual causation
between macroeconomic innovations and dynamic conditional alpha spreads serves as a core
qualifying condition for more effective factor selection with sound economic rigor and intuition.
This new economic insight is one of our key contributions to modern asset pricing model design.

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 639

Appendix 4 discusses the main similarities and differences between our current study and some
concurrent contributions.
In accordance with the core thesis of Kozak et al. (2018), both fundamental and behavioral
factors can help price the cross-section of average returns. In order to determine whether a specific
factor is fundamental, the econometrician can measure factor covariances or factor premiums
with macroeconomic risk innovations. Vector autoregression (VAR) evidence confirms bilateral
Granger causation between dynamic conditional alphas and macroeconomic risk innovations.
VAR evidence further supports this mutual causation for Fama and French (2015) fundamental
factors with the plausible exceptions of both momentum and partial value. In light of this evidence,
it would be informative to analyze the dynamic conditional alpha time-series (cf. Appendix 1
time-series visualization of dynamic conditional alphas). When each dynamic conditional alpha
is nil on average but can be persistently positive and negative during different phases of the real
business cycle, the conditional moments and factor premiums can help inform empirical asset
pricing model design and performance.
Our dynamic analysis of core conditional factor premiums from the Fama and French (2015)
model proposes raising the hurdle for the conventional asset-pricing test. This recommendation
serves as the time-series equivalent to the cross-sectional thesis of Harvey et al. (2016). Also, our
prime empirical analysis contradicts McLean and Pontiff’s (2016) recent conjecture that academic
research partially erodes stock return predictability because investors are able to learn from a vari-
ety of anomalies. To the extent that the dynamic conditional factor premiums exhibit substantial
volatility over time, the vast majority of dynamic conditional alphas are not significantly far from
zero while there is no sufficient evidence to reject the hypothesis that our conditional factor model
carries a correct specification. For this reason, we can reconcile most ubiquitous anomalies with a
dynamic variant of the Fama-French factor model. Also, our study offers a mild refutation of Berk
and Van Binsbergen’s (2016) and Barber et al. (2016) overall qualitative conclusion that the CAPM
is the clear victor in the horserace against the other dynamic-equilibrium and factor models. In
contrast, our work shows that the dynamic conditional factor model provides progress toward a
positive portrayal of the quantitative nexus between average return and risk. Our dynamic con-
ditional factor model helps draw a distinction between both rational-risk and behavioral theories
of average return evolution because bilateral causation between macroeconomic innovations and
dynamic conditional alpha spreads shines unique light on the marginal investor’s fundamental
news and economic expectations about the cross-section of average returns. This insight offers
economic logic, rigor, and intuition for our empirical analysis in response to the recent landmark
statistical discovery of Kozak et al. (2018). As a consequence, the myriad contributions of our
current study help address at least part of the concern and suspicion in the recent reappraisals of
asset pricing model tests (cf. Lewellen et al. (2010); Berk and van Binsbergen (2016); Barber et al.
(2016); Harvey et al. (2016); Fama and French (2015, 2016); Kozak et al. (2017, 2018)).

5. Conclusion
In response to Kozak et al. (2018) recent critique of many horseraces among factor models, we
apply a new approach to addressing at least part of the concern and suspicion in several reap-
praisals of asset pricing tests (cf. Lewellen et al. (2010); Berk and van Binsbergen (2016); Barber
et al. (2016); Harvey et al. (2016); Fama and French (2015, 2016); Kozak et al. (2017, 2018)).
We extract dynamic conditional factor premiums from the Fama and French (2015) model and
then find that most anomalies disappear after one accounts for time variation in these premi-
ums. Mutual causation between dynamic conditional alpha spreads and macroeconomic surprises
serves as a core qualifying condition for relevant fundamental factor selection with sound eco-
nomic rigor and motivation. To the extent that macroeconomic innovations manifest in the form
of dynamic conditional alphas, this causation reveals the marginal investor’s fundamental news
and macroeconomic expectations about the cross-section of average returns.

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 640 M.-W. Hung and A. J.-Y. Yeh

Specifically, our evidence bolsters the ubiquitous use of Fama and French (2015) factors that
reflect the marginal investor’s response to fundamental news about the cross-section of average
returns. Our econometric results lend credence to Fama and French’s (1996, 2004, 2008, 2015,
2016) perennial reluctance to encompass Carhart (1997) momentum in their factor model. In a
conceptual domain, we link the dynamic conditional factor model results to recent advances in
the intertemporal CAPM context (Merton (1973); Campbell (1993); Campbell and Vuolteenaho
(2004); Campbell et al. (2017)). Overall, our current study serves as an incremental step toward
better deciphering a distinction between the rational risk and behavioral mispricing paradigms.
The ARMA-GARCH conditional mean-variance specifications and vector autoregressions
(VAR) are important in the current study. The ARMA-GARCH part is important because we
can apply these conditional mean-variance models to yield dynamic conditional alpha forecasts
to help predict future macroeconomic fluctuations. Further, the VAR part is important because
we can establish mutual causation between dynamic alphas and macro innovations. In turn, this
mutual causation serves as a core qualifying condition for valid and effective fundamental fac-
tor selection. In combination, the resultant econometric tests and results bolster our empirical
response to the recent critique by Kozak et al. (2017 and 2018): the fundamental factors would
covary with macroeconomic fluctuations (but the behavioral mispricing factors would not). As a
consequence, our empirical work helps draw a distinction between the rational risk and behavioral
strands of macroeconomic assets return prediction.
We agree with Kozak et al. (2018) qualitative conclusion on “observational equivalence
between most rational risk factor models and behavioral mispricing factor models”. As we have
discussed in the introduction of the current study, this consensus resonates with the relentless
debate and elusive quest of a factor model for our profession. These advances shed new light on the
major essential need for financial economists to “[develop-and-test] structural asset-pricing mod-
els with specific assumptions about investor beliefs and preferences that can deliver testable pre-
dictions about (1) the fundamental factors that should be in the [stochastic discount factor] (SDF),
and (2) the probability distributions under which this SDF prices assets” (Kozak et al. 2018).
A concurrent contribution moves in this direction. Yeh (2021) derives and estimates the SDF
from financial intermediary capital strength and then uses the Euler, asset return prediction, and
price multiple valuation equations to test stock market anomaly persistence over a broad basket
of long-short extreme decile strategies. Yeh (2021) further tests the empirical predictions of the
dynamic stochastic general equilibrium (DSGE) structural model to find evidence in support of
his permanent capital hypothesis. This new frontier shows some promise in applying “structural
macrofinance models” of both investor beliefs and preferences for economists to make progress
in macroeconomic asset return prediction.

Visualization: Dynamic conditional alphas for the extreme deciles

This visualization shows the time-series charts of the Fama-French dynamic conditional alphas
for the extreme deciles for the stock portfolio tilts such as size, value, momentum, profitabil-
ity, investment, short-run return reversal, and long-run return reversal. These dynamic factor
premiums exhibit wide variation for the top and bottom deciles. Some large comovements tend
to influence the central tendency of the alpha spread between the extreme deciles. Each alpha
spread often switches its sign and becomes econometrically insignificant. In addition to Gibbons
et al. (1989) AGRS F-test, the AGMM C-test and AGMM Q-test of dynamic multifactor mean-
variance efficiency (MMVE) suggests that the average alpha spreads are not different from zero.
The distance between the squared Sharpe ratios for each individual stock portfolio and the MMVE
tangency portfolio is not large enough for one to reject the null hypothesis of a correct asset pricing
model specification. This inference accords with the spirit of the intertemporal context of Merton
(1973), Campbell (1993), and Fama (1996). Investors care about not only their terminal wealth
but also several behavioral considerations such as human capital, labor income, consumption, and
hedging investment opportunities that covary with the conditional expectations of their terminal
wealth. In this light, the Fama and French (2015) factors serve as valid and relevant state variables

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 641

that reflect these co-movements in response to the typical investor’s demand for hedging instru-
ments. To the extent that the dynamic factor premiums on each Fama and French (2015) state
variable is econometrically significant across the entire data span (i.e. each dynamic multifactor
beta consistently differs from zero), the resultant dynamic alpha exhibits too much variability for
the pricing error to be significant enough for the econometrician to reject the null hypothesis that
the dynamic multi-factor model specification is correct.

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 642 M.-W. Hung and A. J.-Y. Yeh

Supplementary material. To view supplementary material for this article, please visit http://doi.org/10.1017/
S1365100523000184

Notes
1 In our current empirical assessment, we consider a reasonably wide array of ubiquitous asset pricing anomalies such as size
(Banz (1981)), value (Basu (1977); Rosenberg et al. (1985); Fama and French (1992); Fama and French (1998); Lakonishok
et al. (1994), medium-term return momentum (Jegadeesh and Titman (1993, 2001); Chan et al. (1996)), asset investment
growth (Titman et al. (2004); Cooper et al. (2008)), operating profitability (Haugen and Baker (1996); Collins and Hribar
(2000); Dechow et al., (1999); Richardson et al. (2005); Fama and French (2006b); Novy-Marx (2013), and contrarian long-
term return reversal (DeBondt and Thaler (1985); Lakonishok et al. (1994); Fama and French (1996, 1998)). Fama and French
(2004) provide a meticulous survey of the main anomalies that point to the empirical embarrassment of the CAPM. Fama
and French (2008) revisit the empirical assessment of these key anomalies and in turn suggest that these anomalies tend
to concentrate in the extreme deciles or in the microcap portfolio. Some more recent studies replicate a broader basket of
anomalies (cf. Fama and French (2016); Harvey et al. (2016); Harvey (2017); Hou et al. (2017); Chordia et al. (2017)). Our
primary and ultimate goal is not to compete with these prominent authors with more replication. Instead, the current study
seeks to establish “mutual causation” between dynamic conditional alphas and macro surprises as a core qualifying condition
for relevant and effective factor selection in subsequent asset pricing model design.
2 The conditional asset pricing literature can be traced back to the econometric contributions of Harvey (1989), Shanken
(1990), Jagannathan and Wang (1996), Lettau and Ludvigson (2001). Ferson and Harvey (1991, 1993, 1999) empirically link
multifactor betas to economic fluctuations. Several studies point out the importance of identifying the correct and relevant
set of state variables (Harvey, 1989; Shanken, 1990; Jagannathan and Wang, 1996; Cochrane, 2001: 145). Lewellen and Nagel
(2006) avoid this problem by using short-window regressions. Furthermore, Ang and Chen (2007) and Fama and French
(2006) both assume some particular structure of auto-regressive mean reversion or structural breaks in the time-series behav-
ior of market beta. Adrian and Franzoni (2009) allow market beta to vary over time with a univariate version of the recursive
filter that the current paper proposes in the multivariate context. Ang and Kristensen (2012) test the conditional CAPM and
the conditional Fama-French three-factor model and report evidence in favor of the alternative hypothesis that the pricing
errors are too large for the conditional model to be correctly specified.
3 There are several main differences between the current paper and the previous studies of conditional factor models. First,
we consider all of Fama and French’s (2015) fundamental factors in the recursive estimation of dynamic conditional factor
premiums in contrast to the narrower focus on the single-beta CAPM (Lewellen and Nagel (2006); Fama and French (2006a);
Petkova and Zhang (2005); Ang and Chen (2007); Adrian and Franzoni (2009) and the prior Fama-French factor model (Ang
and Kristensen (2012)). Should the model exclude some relevant state variables, the incorrect specification would lead to
an inconsistent estimator and also would produce significant alphas. In effect, each alpha absorbs the sum product of each
dynamic factor premium and the corresponding state variable that one excludes from the model. Therefore, the inclusion
of all of the Fama and French (2015) return spreads captures a broader gamut of state variables. This information set can
more accurately span the mean-variance space. In turn, the consistent estimation of dynamic conditional factor premiums
minimizes the likelihood of omitted-variables bias. Second, the current study uses a dynamic version of the Fama and French
(2015) factor model to assess the persistence of anomalies such as size, value, momentum, asset investment growth, operating
profitability, short-run return reversal, and long-term return reversal. Unlike Petkova and Zhang (2005), Lewellen and Nagel
(2006), Fama and French (2006a), and Ang and Kristensen (2012), we find evidence in favor of the null hypothesis that a
dynamic multifactor model adequately explains the long-short return spreads from portfolio tilts that consistently generate
static anomalous returns. All of these anomalies are not robust after we consider the dynamic nature of conditional factor
premiums. Third, our dynamic application is simple, requires no stringent time-series structure, and retains parsimony and
flexibility in econometric usage. Unlike Ang and Chen (2006), we specify no highly parameterized latent-variable process
to characterize the evolution of conditional factor premiums (Fama and French (2006a: 2177)). Unlike Lewellen and Nagel
(2006), Fama and French (2006a), and Ang and Kristensen (2012), we set no particular time interval for updating the estima-
tion of conditional factor premiums. Neither do we apply any specific kernel method or bandwidth choice to estimate smooth
conditional factor premiums (Ang and Kristensen (2012)). Time variation in factor premiums is not conditional on macroe-
conomic fluctuations. The current study differs from the prior studies that assess time variation in macroeconomic factor
premiums (cf. Ferson and Harvey (1991, 1993, 1999); Lewellen and Nagel (2006); Ang and Kristensen (2012). However, each
recursively fit dynamic factor premium can be generalized as a common financial time series that contains rich and mean-
ingful economic content. It is plausible for conditional factor premiums to convey useful information about macroeconomic
trends and cycles.
4 There are a couple of reasons for this model comparison. First, the vast majority of dynamic conditional alphas are sta-
tistically insignificant so that most of the pricing errors or alphas are near zero in the alternative dynamic conditional
specification. Second, the new conditional specification test results prevail in favor of the alternative hypothesis that only the
dynamic conditional estimator is consistent. This latter reason highlights the power of measurement error minimization that
the recursive multivariate filter attains for consistent estimation. Overall, the dynamic conditional factor model outperforms

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 643

its static counterpart in light of the conditional specification test evidence and dynamic conditional alpha insignificance. In
effect, both the recursive multivariate filter and conditional specification test contribute to the econometric toolkit for empiri-
cal asset pricing analysis. Not only does this advancement pose a new conceptual challenge to the conventional use of ordinary
least-squares (OLS) Fama-French time-series regressions for empirical asset pricing analysis, but this econometric innovation
also suggests that dynamic conditional alpha estimation serves as a novel and useful quantitative method for a broad vari-
ety of financial applications such as cost-of-equity-capital estimation, corporate event study, financial risk management, and
mutual fund performance evaluation.
5 We are grateful to the Associate Editor for his comments on this aspect of our vector autoregression analysis.

References
Adrian, T. and F. Franzoni. (2009) Learning about beta: Time-varying factor loadings, expected returns, and the conditional
CAPM. Journal of Empirical Finance 16(4), 537–556.
Anderson, C. W. and L. Garcia-Feijoo. (2006) Empirical evidence on capital investment, growth options, and security returns.
Journal of Finance 61(1), 171–194.
Ang, A. and J. Chen. (2007) CAPM over the long run. Journal of Empirical Finance 14(1), 1–40.
Ang, A. and D. Kristensen. (2012) Testing conditional factor models. Journal of Financial Economics 106(1), 132–156.
Asness, C. S., T. J. Moskowitz and L. H. Pedersen. (2013) Value and momentum everywhere. Journal of Finance 68(3),
929–985.
Avramov, D. and T. Chordia. (2006) Asset pricing models and financial market anomalies. Review of Financial Studies 19(3),
1001–1040.
Baker, M. and J. Wurgler. (2007) Investor sentiment in the stock market. Journal of Economic Perspectives 21(2), 129–152.
Baker, S. R., N. Bloom and S. J. Davis. (2016) Measuring economic policy uncertainty. Quarterly Journal of Economics 131(4),
1593–1636.
Banz, R. W. (1981) The key relationship between return and market value of common stocks. Journal of Financial Economics
9(1), 3–18.
Barber, B. M., X. Huang and T. Odean. (2016) Which factors matter to investors? Evidence from mutual fund flows. Review
of Financial Studies 29(10), 2600–2642.
Barberis, N., A. Shleifer and J. Wurgler. (2005) Comovement. Journal of Financial Economics 75(2), 283–317.
Barillas, F. and J. Shanken. (February 2018) Comparing asset pricing models. Journal of Finance 73(2), 715–754.
Basu, S. (1977) Investment performance of common stocks in relation to their price-earnings ratios: a test of the efficient
market hypothesis. Journal of Finance 32(3), 663–682.
Berk, J. and J. Van Binsbergen. (2016) Assessing asset pricing models using revealed investor preference. Journal of Financial
Economics 119(1), 1–23.
Berk, J. B. and R. C. Green. (2004) Mutual fund flows and performance in rational markets. Journal of Political Economy
112(6), 1269–1295.
Berk, J. B., R. C. Green and V. Naik. (1999) Optimal investment, growth options, and security returns. Journal of Finance
54(5), 1553–1607.
Black, F. (1986) Noise. Journal of Finance 41(3), 529–543.
Bollerslev, T. (1986) Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics 31(3), 307–327.
Brennan, M. J., T. Chordia and A. Subrahmanyam. (1998) Alternative factor specifications, security characteristics, and the
cross-section of expected stock returns1. Journal of Financial Economics 49(3), 345–373.
Campbell, J. (1993) Intertemporal asset pricing without consumption data. American Economic Review 83(3), 487–512.
Campbell, J., S. Giglio, C. Polk and R. Turley. (2017) An intertemporal CAPM with stochastic volatility. Journal of Financial
Economics 128(2), 207–233.
Campbell, J., J. Hilscher and J. Szilagyi. (2008) In search of distress risk. Journal of Finance 63(6), 2899–2939.
Campbell, J., A. W. Lo and A. C. MacKinlay. (1997) The Econometrics of Financial Markets. Princeton, NJ: Princeton
University Press.
Campbell, J., C. Polk and T. Vuolteenaho. (2009) Growth or glamour? Fundamentals and systematic risk in stock returns.
Review of Financial Studies 23(1), 305–344.
Campbell, J. and T. Vuolteenaho. (2004) Bad beta, good beta. American Economic Review 94(5), 1249–1275.
Carhart, M. M. (1997) On persistence in mutual fund performance. Journal of Finance 52(1), 57–82.
Carlson, M., A. Fisher and R. Giammarino. (2004) Corporate investment and asset price dynamics: Implications for the
cross-section of returns. Journal of Finance 59(6), 2577–2603.
Carlson, M., A. Fisher and R. Giammarino. (2006) Corporate investment and asset pricing dynamics: Implications for SEO
event studies and long-run stock return performance. Journal of Finance 61(3), 1009–1034.
Chan, L. K. C., N. Jegadeesh and J. Lakonishok. (1996) Momentum strategies. Journal of Finance 51(5), 1681–1713.
Chen, L., R. Novy-Marx and L. Zhang. (2011) An alternative three-factor model.
Chordia, T., A. Goyal and A. Saretto. (2017) p-hacking: Evidence from 2 million trading strategies.

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 644 M.-W. Hung and A. J.-Y. Yeh

Cochrane, J. (2005) Asset Pricing. Princeton, NJ: Princeton University Press.

Colacito, R. and M. Croce. (2011) Risks for the long run and the real exchange rate. Journal of Political Economy 119(1),
153–181.
Colacito, R. and M. Croce. (2013) International asset pricing with recursive preferences. Journal of Finance 68(6), 2651–2686.
Collins, D. W. and P. Hribar. (2000) Earnings-based and accrual-based market anomalies: One effect or two? Journal of
Accounting and Economics 29(1), 101–123.
Cooper, I. (2006) Asset pricing implications of non-convex adjustment costs and irreversibility of firm investment. Journal of
Finance 61(1), 139–170.
Cooper, M. J., H. Gulen and M. J. Schill. (2008) Asset growth and the cross-section of stock returns. Journal of Finance 63(4),
1609–1651.
Daniel, K., D. Hirshleifer, L. Sun, K. Daniel, L. Mota, S. Rottke and T. Santos. (2017) Short and long horizon behavioral
factors. Review of Financial Studies 33(4), 1673–1736.
Daniel, K., L. Mota, S. Rottke and T. Santos. (2017) The cross-section of risk and return.
Daniel, K. and S. Titman. (1997) Evidence on the characteristics of cross-sectional variation in stock returns. Journal of
Finance 52(1), 1–33.
Davis, J. L., E. F. Fama and K. R. French. (2000) Characteristics, covariances, and average returns: 1929 to 1997. Journal of
Finance 55(1), 389–406.
DeBondt, W. F. M. and R. Thaler. (1985) Does the stock market overreact? Journal of Finance 40(3), 793–805.
Dechow, P. M., A. P. Hutton and R. G. Sloan. (1999) An empirical assessment of the residual income valuation model. Journal
of Accounting and Economics 26(1), 1–34.
Donangelo, A. (2014) Labor mobility: Implications for asset pricing. Journal of Finance 69(3), 1321–1346.
Durbin, J. (1954) Errors in variables. Review of the International Statistical Institute 1(3), 23–32.
Eiling, E. (2013) Industry-specific human capital, idiosyncratic risk, and the cross-section of expected stock returns. Journal
of Finance 68(1), 43–84.
Engle, R. (1982) Autoregressive conditional heteroskedasticity with estimates of the variance of United Kingdom inflation.
Econometrica 50(4), 987–1007.
Epstein, L. G. and S. E. Zin. (1989) Substitution, risk aversion, and the temporal behavior of consumption and asset returns:
A theoretical framework. Econometrica 57(4), 937–969.
Fama, E. F. (1996) Multifactor portfolio efficiency and multifactor asset pricing. Journal of Financial and Quantitative Analysis
31(4), 441–465.
Fama, E. F. and K. R. French. (1992) The cross-section of expected stock returns. Journal of Finance 47(2), 427–465.
Fama, E. F. and K. R. French. (1993) Common risk factors in the returns on stocks and bonds. Journal of Financial Economics
33(1), 3–56.
Fama, E. F. and K. R. French. (1995) Size and book-to-market factors in earnings and returns. Journal of Finance 50(1),
131–155.
Fama, E. F. and K. R. French. (1996) Multifactor explanations of asset pricing anomalies. Journal of Finance 51(1), 55–84.
Fama, E. F. and K. R. French. (1998) Value versus growth: the international evidence. Journal of Finance 53(6), 1975–1999.
Fama, E. F. and K. R. French. (2004) The capital asset pricing model: Theory and evidence. Journal of Economic Perspective
18(3), 25–46.
Fama, E. F. and K. R. French. (2006a) The value premium and the CAPM. Journal of Finance 61(5), 2163–2185.
Fama, E. F. and K. R. French. (2006b) Profitability, investment, and average returns. Journal of Financial Economics 82(3),
491–518.
Fama, E. F. and K. R. French. (2008) Dissecting anomalies. Journal of Finance 63(4), 1653–1678.
Fama, E. F. and K. R. French. (2015) A five-factor asset pricing model. Journal of Financial Economics 116(1), 1–22.
Fama, E. F. and K. R. French. (2016) Dissecting anomalies with a five-factor model. Review of Financial Studies 29(1), 69–103.
Ferson, W. and C. Harvey. (1991) The variation of economic factor premiums. Journal of Political Economy 99(2),
385–415.
Ferson, W. and C. Harvey. (1993) The risk and predictability of international equity returns. Review of Financial Studies 6(3),
527–566.
Ferson, W. and C. Harvey. (1999) Conditioning variables and the cross section of stock returns. Journal of Finance 54(4),
1325–1360.
Gibbons, M., S. Ross and J. Shanken. (1989) A test of the efficiency of a given portfolio. Econometrica 57(5), 1121–1152.
Glosten, L. R., R. Jagannathan and D. E. Runkle. (1993) On the relation between the expected value and the volatility of the
nominal excess return on stocks. Journal of Finance 48(5), 1779–1801.
Gomes, J., L. Kogan and L. Zhang. (2003) Equilibrium cross-section of returns. Journal of Political Economy 111(4), 693–732.
Griffin, J. M. and M. L. Lemmon. (2002) Book-to-market equity, distress risk, and stock returns. Journal of Finance 57(5),
2317–2336.
Hahn, J. and H. Lee. (2006) Yield spreads as alternative risk factors for size and book-to-market. Journal of Financial and
Quantitative Analysis 41(2), 245–269.

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 Macroeconomic Dynamics 645

Harvey, A. and N. Shephard. (1993) Structural time series models.

Harvey, C. (1989) Time-varying conditional covariances in tests of asset pricing models. Journal of Financial Economics 24(2),
289–317.
Harvey, C. and Y. Liu. (2017) Presidential address: the scientific outlook in financial economics. Journal of Finance
Harvey, C. R. and Y. Liu. (2014a) Backtesting.
Harvey, C. R. and Y. Liu. (2014c) Evaluating trading strategies.
Harvey, C. R. and Y. Liu. (2014d) Incremental factors.
Harvey, C. R. and Y. Liu. (2014b) Multiple testing in economics Available at SSRN: https://ssrn.com/abstract=2358214.
Harvey, C. R., Y. Liu and H. Q. Zhu. (2016) ... and the cross-section of expected returns. Review of Financial Studies 29(1),
5–68.
Haugen, R. A. and N. Baker. (1996) Commonality in the determinants of expected stock returns. Journal of Financial
Economics 41(3), 401–439.
Hausman, J. A. (1978) Specification tests in econometrics. Econometrica 46(6), 1251–1271.
He, L. Z. Z., S. W. Huh and B. S. Lee. (2010) Dynamic factors and asset pricing. Journal of Financial and Quantitative Analysis
45(3), 707–737.
Higham, N. J. (2002) Computing the nearest correlation matrix: A problem from finance. IMA Journal of Numerical Analysis
22(3), 329–343.
Hou, K., C. Xue and L. Zhang. (2015) Digesting anomalies: an investment approach. Review of Financial Studies 28(3),
650–705.
Hou, K., C. Xue and L. Zhang. (2017) Replicating anomalies. NBER #23394.
Hung, M. W. (1994) The interaction between non-expected utility and asymmetric market fundamentals. Journal of Finance
49(1), 325–343.
Jagannathan, R. and Z. Wang. (1996) The conditional CAPM and the cross-section of expected returns. Journal of Finance
51(1), 3–53.
Jegadeesh, N. and S. Titman. (1993) Returns to buying winners and selling losers: implications for stock market efficiency.
Journal of Finance 48(1), 65–91.
Jegadeesh, N. and S. Titman. (2001) Profitability of momentum strategies: an evaluation of alternative explanations. Journal
of Finance 56(2), 699–720.
Johnson, T. C. (2002) Rational momentum effects. Journal of Finance 57(2), 585–608.
Kalman, R. (1960) A new approach to linear filtering and prediction problems. Journal of Basic Engineering 82(D), 35–45.
Kozak, S., S. Nagel and S. Santosh. (2017) Shrinking the cross-section. NBER #24070.
Kozak, S., S. Nagel and S. Santosh. (2018) Interpreting factor models. Journal of Finance 73(3), 1183–1223.
Lai, T. L. and H. P. Xing. (2008) Statistical Models and Methods for Financial Markets. Springer Science.
https://www.amazon.com/Statistical-Methods-Financial-Springer-Statistics/dp/0387778268.
Lakonishok, J., A. Shleifer and R. Vishny. (1994)Contrarian investment, extrapolation, and risk. Journal of Finance 49(5),
1541–1578.
Lee, C., A. Shleifer and R. Thaler. (1991) Investor sentiment and the closed-end fund puzzle. Journal of Finance 46(1), 75–109.
Lettau, M. and S. Ludvigson. (2001) Consumption, aggregate wealth, and expected stock returns. Journal of Finance 56(3),
815–849.
Lewellen, J. and S. Nagel. (2006) The conditional CAPM does not explain asset pricing anomalies. Journal of Financial
Economics 82(2), 289–319.
Lewellen, J., S. Nagel and J. Shanken. (2010) A skeptical appraisal of asset pricing tests. Journal of Financial Economics 96(2),
175–194.
Li, E. X., D. Livdan and L. Zhang. (2009) Anomalies. Review of Financial Studies 22(11), 4301–4334.
Liew, J. and M. Vassalou. (2000) Can book-to-market, size, and momentum be risk factors that predict economic growth?
Journal of Financial Economics 57(2), 221–245.
Lin, X. J. and L. Zhang. (2013) The investment manifesto. Journal of Monetary Economics 60(3), 351–366.
Lintner, J. (1965) The valuation of risk assets and the selection of risky investments in stock portfolios and capital budgets.
Review of Economics and Statistics 47(1), 13–37.
Liu, L., T. Whited and L. Zhang. (2009) Investment-based expected stock returns. Journal of Political Economy 117(6),
1105–1139.
Liu, L. and L. Zhang. (2008) Momentum profits, factor pricing, and macroeconomic risk. Review of Financial Studies 21(6),
2417–2448.
Liu, L. and L. Zhang. (2014) A neoclassical interpretation of momentum. Journal of Monetary Economics 67(3), 109–128.
Maddala, G. S., Rao, C. R. and Vinod, H. D.. (eds.) Handbook of Statistics, vol. 11, pp. 261–302.
Markowitz, H. M. (1952) Portfolio selection. Journal of Finance 7(1), 77–91.
McLean, R. D. and J. Pontiff. (2016) Does academic research destroy stock return predictability? Journal of Finance 71(1),
5–23.
Merton, R. (1973) An intertemporal capital asset pricing model. Econometrica 41(5), 867–887.

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press

 646 M.-W. Hung and A. J.-Y. Yeh

Merton, R. (1980) On estimating the expected return to the market: An exploratory investigation. Journal of Financial
Economics 8(4), 323–361.
Nagel, S. and K. J. Singleton. (2011) Estimation and evaluation of conditional asset pricing models. Journal of Finance 66(3),
873–909.
Nakamura, A. and M. Nakamura. (1981) On the relations among several specification error tests presented by Durbin, Wu,
and Hausman. Econometrica 49(6), 1583–1588.
Nelson, D. (1991) Conditional heteroskedasticity in asset returns: A new approach. Econometrica 59(2), 347–370.
Newey, W. K. (1985) Generalized method of moments specification testing. Journal of Econometrics 29(3), 229–256.
Newey, W. K. and K. D. West. (1987) A positive semidefinite, heteroscedasticity-and-autocorrelation consistent covariance
matrix. Econometricia 55(3), 703–708.
Novy-Marx, R. (2013) The other side of value: The gross profitability premium. Journal of Financial Economics 108(1), 1–28.
Novy-Marx, R. and M. Velikov. (2016) A taxonomy of anomalies and their trading costs. Review of Financial Studies 29(1),
104–147.
Petkova, R. (2006) Do the Fama-French factors proxy for innovations in predictive variables? Journal of Finance 61(2),
581–612.
Petkova, R. and L. Zhang. (2005) Is value riskier than growth? Journal of Financial Economics 78(1), 187–202.
Richardson, S., R. Sloan, M. Soliman and I. Tuna. (2005) Accrual reliability, earnings persistence, and stock prices. Journal of
Accounting and Economics 39(3), 437–485.
Rosenberg, B., R. Kenneth and R. Lanstein. (1985) Persuasive evidence of market inefficiency. Journal of Portfolio
Management 11(3), 9–16.
Shanken, J. (1990) Intertemporal asset pricing: An empirical investigation. Journal of Econometrics 45(1-2), 99–120.
Sharpe, W. (1964) Capital asset prices: A theory of market equilibrium under conditions of risk. Journal of Finance 19(3),
425–442.
Sims, C. A. (1980) Macroeconomics and reality. Econometrica, 1–48.
Stambaugh, R. and Y. Yuan. (2016) Mispricing factors. Review of Financial Studies 30(4), 1270–1315.
Tauchen, G. (1985) Diagnostic testing and evaluation of maximum likelihood models. Journal of Econometrics 30(1-2), 415–
443.
Titman, S., J. K. C. Wei and F. X. Xie. (2004) Capital investments and stock returns. Journal of Financial and Quantitative
Analysis 39(4), 677–700.
Tsay, R. S. (2010) Analysis of Financial Time Series. Wiley Publication. https://www.wiley.com/en-us/Analysis+of+Financial+
Time+Series%2C+3rd+Edition-p-9780470414354
Vassalou, M. (2003) News related to future GDP growth as a risk factor in equity returns. Journal of Financial Economics
68(1), 47–73.
Vassalou, M. and Y. H. Xing. (2004) Default risk in equity returns. Journal of Finance 59(2), 831–868.
White, H. (1982) Maximum likelihood estimation of misspecified models. Econometrica 50(1), 1–25.
Wu, D. M. (1973) Alternative tests of independence between stochastic regressors and disturbances. Econometrica 41(4),
733–750.
Yeh, A. J. Y. (2021, Permanent capital hypothesis. PhD dissertation
Zhang, L. (2005) The value premium. Journal of Finance 60(1), 67–103.

Cite this article: Hung M-W and Yeh AJ-Y (2024). “Stock market alphas help predict macroeconomic innovations.”
Macroeconomic Dynamics 28, 612–646. https://doi.org/10.1017/S1365100523000184

https://doi.org/10.1017/S1365100523000184 Published online by Cambridge University Press