Event Studies in Economics and Finance

A. Craig MacKinlay

Journal of Economic Literature, Vol. 35, No. 1. (Mar., 1997), pp. 13-39.

Stable URL:
http://links.jstor.org/sici?sici=0022-0515%28199703%2935%3A1%3C13%3AESIEAF%3E2.0.CO%3B2-W

Journal of Economic Literature is currently published by American Economic Association.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/aea.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org
Sun Jan 27 17:44:51 2008
Journal of Economic Literature
Vol. X X X V (March 1997), pp. 13-39

Event Studies in Economics and

Finance

A. CRAIGMACKINLAY
The IVlzarton Sclzool, University of Pennsylvania
T11nrlk.t to Johri CnrnpDcll, BI-cicc GI-rr~lcly,
A ~ i r l ~ a iLco , ci~lrltico c i ~ ~ o ~ i ~ y ~rcfercc.i,fi)r
~iocii Izel/~fi~I
C O I T I I I I ~ I(i~rcl
I ~ . ~d i < c ~ i ~ < iH~r(v1rc11
o~l. . r ~ r j ~ p o,fro111
rf tlre
Hoclrrry L . \l711itc>C o l t r ~ - f o Fi~icirrcial
r
Kc.rcci~-chir g~-crtefiillycickr~oi~~lcclgerl.

1. Introduction deficit.1 However. amlications in other

I I

fields are also abundant. For example,

E CONOMISTS are frequently asked to
measure the effects of an economic
event on the value of firms. On the sur-
event studies are used in the field of law
and economics to measure tlle impact on
the value of a firm of a change in the
face this seems like a difficult task, but a
regulatory environment (see G. \Villiam
measure can be constructed easily using
Schwert 1981) and in legal liability cases
an event study. Using financial market
event studies are used to assess damages
data, an event study measures the impact
(see Mark Mitchell and Jeffry Netter
of a specific event on the value of a firm.
1994). I n the majority of applications,
The usefulness of such a study comes
the focus is the effect of an event on the
from the fact that, given rationality in
price of a particular class of securities of
the marketplace, the effects of an event
the firm, most often common equity. In
will be reflected immediately in security
this paper the methodology is discussed
prices. Thus a measure of the event's
in terms of applications that use common
economic impact can be constructed
equity. However, event studies can be
using security prices observed over a
applied using debt securities with little
relatively short time period. In contrast,
modification.
direct productivity related measures may
Event studies have a long history. Per-
require many lnontlls or even years of
haps the first published study is James
observation.
Dolley ( 1 9 3 3 ) . In this work, he examines
Tlle event study llas many applica-
the price effects of stock splits, studying
tions. In accounting and finance re-
nominal price changes at the time of the
search, event studies have been applied
split. Using a sample of 9Fj splits from
to a variety of firm specific and economy
1921 to 1931, he finds that the price in-
wide events. Soine examples include
mergers and acquisitions, earnings an-
nouncements, issues of new debt or eq- lT11e first three examples will be discussed later
in the paper. Grant McQueen and Vance Roley
u i t ~ , and announcements of macro- (1993) provide an illustration of the fourth using
economic variables such as the trade n~acroeconon~ic news announcen~ents.
14 Journal of Economic Literature, 1701. XXX17 (March 1997)

creased in 5 7 of the cases and the price with the necessary tools for calculating
declined in only 26 instances. Over the an abnormal return, making statistical in-
decades from the early 1930s until the ferences about these returns, and aggre-
late 1960s the level of sophistication of gating over inany event observations.
event studies increased. John H . Myers The null hypothesis that the event has no
and Archie Bakay (1948), C . Austin impact on the distribution of returns is
Barker (1956, 1957, 1958), and John maintained in Sections 4 and 5. Section 6
Ashley (1962) are examples of studies discusses modifying this null hypotllesis
during this tiine period. The iinprove- to focus only on the inean of the return
ments included removing general stock distribution. Section 7 presents analysis
inarket price inoveinents and separating of the power of an event study. Section 8
out confounding events. In the late presents nonparainetric approaches to
1960s seminal studies by Ray Ball and event studies which eliminate the need
Philip Brown (1968) and Eugene Faina for parametric structure. In some cases
et al. (1969) introduced the methodology theory provides hypotheses concerning
that is essentially the same as that whicll the relation between the magnitude of
is in use today. Ball and Brown consid- the event abnormal return and firm char-
ered the inforination content of earn- acteristics. Section 9 presents a cross-
ings, and Faina e t al. studied the effects sectional regression approach that is use-
of stock splits after removing the effects ful to investigate such hypotheses.
of simultaneous dividend increases. Section 10 considers soine further issues
In the years since these pioneering relating event study design and the pa-
studies, a number of inodifications have per closes with the concluding discussion
been developed. These modifications re- in Section 11.
late to coinplications arising from viola-
tions of the statistical assuinptions used 2. Procedure for an Event Study
in the early work and relate to adjust-
ments in the design to accommodate At the outset it is useful to briefly dis-
more specific hypotheses. Useful papers cuss the structure of an event study. This
whicll deal with the practical iinportance will provide a basis for the discussion of
of inany of the coinplications and adjust- details later. \Vhile there is no unique
inents are the work by Stephen ~ r o w n structure, there is a general flow of
and Jerold Warner published in 1980 and analysis. This flow is discussed in this
1985. The 1980 paper considers imple- section.
mentation issues for data sampled at a The initial task of conducting an event
monthly interval and the 1985 paper study is to define the event of interest
deals with issues for daily data. and identify the period over which the
In this paper, event study inethods are security prices of the firms involved in
reviewed and suminarized. The paper this event will be examined-the event
begins with discussion of one possible window. For example, if one is looking at
procedure for conducting an event study the information content of an earnings
in Section 2. Section 3 sets up a sample with daily data, the event will be the
event study which will be used to illus- earnings announcement and the event
trate the inethodology. Central to an window will include the one day of the
event study is the ineasureinent of an ab- announcement. I t is custoinary to define
normal stock return. Section 4 details the event window to be larger than the
the first step-measuring the norinal specific period of interest. This perinits
performance-and Section 5 follows examination of periods surrounding the
 M a c K i n l a y : E v e n t S t u d i e s i n E c o n o m i c s a n d Finance 15

event. In practice, the period of interest clloices for modeling the normal re-
is often expanded to multiple days, in- turn-the constant mean return model
cluding at least tlle day of tlle an- where X , is a constant, and the market
nouncement and the day after the an- model where XiT is the market return.
nouncement. This captures the price The constant mean return model, as the
effects of an,nounceinents which occur name implies, assumes that the mean
after the stock market closes on the an- return of a given security is constant
nouncement day. The periods prior to through time. The market model as-
and after the event may also be of inter- sumes a stable linear relation between
est. For example, in the earnings an- the market return and the security re-
nouncement case, tlle market may ac- turn.
quire information about the earnings Given the selection of a norlnal perfor-
prior to the actual announcement and mance model, tlle estimation window
one can investigate this possibility by ex- needs to be defined. The most common
amining pre-event returns. choice, when feasible, is using the period
After identifying the event, it is neces- prior to the event window for tlle estima-
sary to determine tlle selection criteria tion window. For example, in an event
for the inclusion of a given firm in the study using daily data and tlle market
study. The criteria may involve restric- model, the market model parameters
tions imposed by data availability such as could be estimated over the 120 days
listing on the New York Stock Exchange prior to the event. Generally the event
or the American Stock Exchange or may period itself is not included in the esti-
involve restrictions such as membership mation period to prevent the event from
in a specific industry. At this stage it is influencing the norinal performance
useful to sumlnarize some sainple char- model parameter estimates.
acteristics (e.g., firm market capitaliza- With the parameter estimates for the
tion, industry representation, distri- normal performance model, the abnor-
bution of events througll time) and note mal returns can be calculated. Next
any potential biases which may have comes the design of the testing frame-
been introduced through the sample se- work for tlle abnormal returns. Impor-
lection. tant considerations are defining the null
Appraisal of the event's impact re- hypotllesis and determining the tech-
quires a measure of the abnormal return. niques for aggregating tlle individual
The abnormal return is the actual ex post firm abnormal returns.
return of the security over the event win- The presentation of the empirical re-
dow minus the norinal return of the firin sults follows the formulation of the
over the event window. The normal re- econometric design. In addition to pre-
turn is defined as the expected return senting the basic einpirical results, the
without conditioning on the event taking presentation of diagnostics can be fruit-
place. For firm i and event date T the ful. Occasionally, especially in studies
abnormal return is with a limited number of event observa-
tions, tlle empirical results can be heav-
ARi, = R,, - E(R,,IX',) (1) ily influenced by one or two firms.
where AR,,, R,,, and E(R,,lX',) are the ab- Knowledge of this is important for gaug-
normal, actual, and normal returns re- ing the importance of the results.
spectively for time period T. X', is the Ideally the empirical results will lead
conditioning information for the norlnal to insights relating to understanding the
return model. There are two common sources and causes of the effects (or lack
16 Journal of Econonzic Literature, Vol. XXXV (March 1997)

of effects) of the event under study. Ad- is Datastreain, and the source of the ac-
ditional analysis inay be included to dis- tual earnings is Compustat.
tinguish between competing explana- If earnings announcements convey in-
tions. Concluding coiilinents complete formation to investors, one would expect
the study. the announcement impact on the mar-
ket's valuation of the firin's equity to de-
3. An Exan~pleof nrz Event Study pend on the magnitude of the unex-
pected coinponent of the announcement.
The Financial Accounting Standards Thus a ineasure of the deviation of the
Board (FASB) and the Securities Ex- actual announced earnings from the mar-
change Coininission strive to set report- ket's prior expectation is required. For
ing regulations so that financial state- constructing such a ineasure, the mean
inents and related inforination releases quarterly earnings forecast reported by
are informative about the value of the the Institutional Brokers Estimate Sys-
firm. In setting standards, the informa- ten1 (I/B/E/S) is used to proxy for the
tion content of the financial disclosures market's expectation of earnings. I/B/E/S
is of interest. Event studies provide an compiles forecasts from analysts for a
ideal tool for examining the inforn~ation large nuinber of companies and reports
content of the disclosures. suminary statistics each month. Tlle
In this section the description of an mean forecast is taken froin the last
exainple selected to illustrate event inontll of the quarter. For exainple, the
study inetllodology is presented. One mean third quarter forecast froin Sep-
particular type of disclosure-quarterly tember 1990 is used as the measure of
earnings announcements-is considered. expected earnings for the third quarter
The objective is to investigate the infor- of 1990.
ination content of these announce- To facilitate the exainination of the
ments. In other words, the goal is to see impact of the earnings announceinent on
if the release of accounting inforniation the value of the firm's equity, it is essen-
provides inforniation to the marketplace. tial to posit the relation between the in-
If so there should be a correlation be- formation release and the change in
tween the observed change of the mar- value of tlie equity. In this example the
ket value of the company and the infor- task is straightforward. If the earnings
mation. disclosures have information content,
The example will focus on the quar- higlier than expected earnings sllould be
terly earnings announcements for the 30 associated with increases in value of the
firins in the Dobv Jones Industrial Index equity and lower than expected earnings
over the five-year period froin January with decreases. To capture this associa-
1989 to December 1993. These an- tion, each announcement is assigned to
nouncements correspond to the quar- one of three categories: good news, no
terly earnings for the last quarter of 1988 news, or bad news. Each announceinent
through the third quarter of 1993. Tlle is categorized using the deviation of the
five years of data for 30 firms provide a actual earnings froin the expected earn-
total sainple of 600 annouiicenients. For ings. If the actual exceeds expected by
each firm and quarter, three pieces of in- inore thaii 2.5 percent the announce-
formation are compiled: the date of the ment is designated as good news, and if
announcement, the actual earnings, and the actual is inore than 2.5 percent less
a ineasure of the expected earnings. The than expected the announceinent is des-
source of the datc of the announcement ignated as bad news. Those announce-
 MacKinlay: E v e n t Studies i n Economics a n d Finance 17
inents where the actual earnings is in the For the statistical inodels, the assump-
5 percent range centered about the ex- tion that asset returns are jointly multi-
pected earnings are designated as no variate normal and independently and
news. Of the 600 announcements, 189 identically distributed tllrougll tiine is
are good news, 173 are no news, and the imposed. This distributional assuinption
remaining 238 are bad news. is sufficient for the constant mean return
With the announcements categorized, model and the market model to be cor-
the next step is to specify the parameters rectly specified. While this assuinption is
of the empirical design to analyze the eq- strong, in practice it generally does not
uity return, i.e., the percent change in lead to probleins because the assuinption
value of the equity. I t is necessary to is empirically reasonable and inferences
specify a leiigtll of observation interval, using the normal return inodels tend to
an event window, and an estimation win- be robust to deviations from the assump-
dow. For this exainple the interval is set tion. Also one can easily inodify the sta-
to one day, thus daily stock returns are tistical framework so that the analysis of
used. A 41-day event window is em- the abnormal returns is autocorrelation
ployed, comprised of 20 pre-event days, and lieteroskedasticity consistent by us-
the event day, and 20 post-event days. ing a generalized method-of-moinents
For each announcement the 250 trading approach.
day period prior to the event window is
used as tlle estirnatioii window. After A. Constant Mean Return Model
presenting the methodology of an event
Let p, be the mean return for asset i .
study, this exainple will be drawn upon
Then the constant mean return model is
to illustrate the execution of a study.
Rir = CLi + <it (2)
4. Models for Measuring Normal E(<ii) = 0 var = G!.
Performance
where R,t is the period-t return on secu-
A nuinber of approaches are available rity i and ( , t is tlle time period t distur-
to calculate the normal return of a given bance term for security i with an expec-
security. The approaclles can be loosely tation of zero and variance oil.
grouped into two categories-statistical Altllougli the constant mean return
and econoinic. Models in the first cate- inodel is perhaps tlie simplest inodel,
gory follow from statistical assuinptions Brown and Warner (1980, 1985) find it
concerning the behavior of asset returns often yields results similar to those of
and do not depend on any economic ar- inore sophisticated iiiodels. This lack of
guments. I n contrast, inodels in the sec- sensitivity to the model can be attributed
ond category rely on assuinptions con- to the fact that the variance of the abnor-
cerning investors' behavior and are not mal return is frequently not reduced
based solely on statistical assumptions. I t iiiucl~by choosing a more sophisticated
should, however, be noted that to use model. When using daily data the model
econoinic models in practice it is neces- is typically applied to nominal returns.
sary to add statistical assuinptions. Thus With inonthly data the model can be ap-
the potential advantage of economic plied to real returns or excess returns
inodels is not the absence of statistical (tlie return in excess of tlle nominal risk
assumptions, but the opportunity to cal- free return generally measured using the
culate more precise measures of the nor- U.S. Treasury Bill with one month to
inal return using economic restrictions. maturity) as well as nominal returns.
1S Journal of Econonzic Literature, Val. XXXV ( M a r c h 1997)

B . Market Model The market model is an examnle L

of a one
factor model. Other inultifactor models
Tlle inarket model is a statistical
include industry indexes in addition to
inodel wllicll relates the return of any
the market. \Villiam Sllarpe (1970) and
given security to the return of the inar-
Sharpe, Gordon Alexander, and Jeffery
ket portfolio. The model's linear specifi-
Bailey (1995, p. 303) provide discussion
cation follows from the assumed joint
of index models with factors based on in-
norinality of asset returns. For any secu-
dustry classification. Another variant of a
rity i the inarket inodel is
factor model is a procedure wllicll calcu-
lates the abnormal return by taking the
difference between the actual return and
a portfolio of firms of siinilar size, where
size is measured by market value of eq-
where R,, and R,,,t are the period-t re- uity. In this approach typically ten size
turns on security i and the inarket port- groups are considered and the loading on
folio, respectively, and ~ , ist the zero
the size portfolios is restricted to unity.
mean disturbance term. a,, P,, and o$, This procedure implicitly assumes that
are the parameters of the market inodel. expected return is directly related to
In applications a broad based stock in- inarket value of equity.
dex is used for the inarket portfolio, Generally, the gains from einploying
with the S&P 500 Index, the CRSP inultifactor inodels for event studies are
Value Weighted Index, and the CRSP limited. The reason for the limited gains
Equal Weighted Index being popular is the empirical fact that the marginal
choices. explanatory power of additional factors
The inarket inodel represents a poten- the inarket factor is sinall, and hence,
tial iinproveinent over the constant mean there is little reduction in the variance of
return model. By removing the portion tlle abnormal return. The variance re-
of the return that is related to variation duction will typically be greatest in cases
in the market's return, the variance of where the sample firins have a coininon
the abnormal return is reduced. This in characteristic, for exainple they are all
turn can lead to increased ability to de- inembers of one industry or they are all
tect event effects. Tlle benefit from us- firins concentrated in one inarket capi-
ing the inarket inodel will depend upon talization group. In these cases tlle use
the R b f the inarket inodel regression. of a inultifactor inodel warrants consid-
The lligller the Rqthe greater is the vari- eration.
ance reduction of the abnormal return, The use of other models is dictated by
and the larger is the gain. data availability. An example of a normal
performance return model implemented
C . Other Statistical Models
in situations with limited data is the mar-
A number of other statistical inodels ket-adjusted return model. For some
have been proposed for modeling the events it is not feasible to have a pre-
norinal return. A general type of statisti- event estimation period for the norinal
cal inodel is the factor model. Factor inodel parameters, and a inarket-ad-
models are nlotivated by the benefits of justed abnorinal return is used. The inar-
reducing the variance of the abnormal ket-adjusted return inodel can be viewed
return by explaining inore of the vari- as a restricted inarket inodel with a, con-
ation in the norinal return. Typically the strained to be zero and p, constrained to
factors are portfolios of traded securities. be one. Because the inodel coefficients
 MacKinlny: Euent Studies i n Economics a n d Finance 19

are prespecified, an estiination period is motivated by the Arbitrage Pricing

not required to obtain parameter esti- Theory. A general finding is that with
mates. An exanlple of wllen such a model the APT the most iinportant factor be-
is used is in studies of the under pricing haves like a market factor and additional
of initial public offerings. Jay Ritter factors add relatively little explanatory
(1991) presents sucll an example. A gen- power. Thus the gains from using an
eral recommendation is to only use such APT motivated model versus the lnarket
restricted models if necessary, and if inodel are small. See Stephen Brown
necessary, consider the possibility of bi- and Mark \Veinstein (1985) for further
ases arising from the ilnposition of the discussion. The main potential gain
restrictions. from using a model based on the arbi-
trage pricing theory is to eliminate the
D . Economic Models
biases introduced by using the CAPM.
Econolnic models can be cast as re- However, because the statistically moti-
strictions on the statistical models to vated lnodels also eliminate these bi-
provide more constrained normal return ases, for event studies such models
models. Two common economic lnodels dominate.
which provide restrictions are the Capi-
tal Asset Pricing Model (CAPM) and the 5. Measuring and Analyzing Abnormal
Arbitrage Pricing Theory (APT). The Returns
CAPM due to Sharpe (1964) and John
Lintner (196Fj) is an equilibrium theory I11 this section the problem of ineasur-
where the expected return of a given as- ing and analyzing abnormal returns is
set is determined by its covariance with considered. The framework is developed
the market portfolio. The APT due to using the market inodel as the norinal
Stephen Ross (1976) is an asset pricing performance return model. The analysis
theory where the expected return of a is virtually identical for the constant
given asset is a linear combination of mean return model.
multiple risk factors. Soine notation is first defined to facili-
The use of the Capital Asset Pricing tate the measurement and analysis of ab-
Model is comnlon in event studies of the normal returns. Returns will be indexed
1970s. However, deviations from the in event time using 7. Defining .t = 0 as
CAPM have been discovered, implying t h e e v e n t d a t e , .t = TI + 1 to 7 = T2 repre-
that the validity of the restrictions im- sents the event window, and 7 = To + 1 to
posed by the CAPM on the lnarket 7 = T I constitutes the estimation window.
inodel is q u e s t i o n a b l e . ~ l l i shas intro- Let L1 = T 1- To and L, = TZ- T1 be the
duced the possibility that the results length of the estimation window and the
of the studies may be sensitive to the event window respectively. Even if the
specific CAPM restrictions. Because event being considered is an an-
this potential for sensitivity can be nouncelnent on given date it is typical to
avoided at little cost by using the market set the event window length to be larger
model, the use of the CAPM has almost than one. This facilitates the use of ab-
ceased. normal returns around the event day in
Similarly, other studies have employed the analysis. \$'hen applicable, the post-
multifactor normal performance models event window will be from .t = T2 + 1 to
T = T; and of length L:i = T:3- T2. The tim-
'Eugene Fama and Kenneth French (1996) ing sequence is illustrated wit11 a time
provide discussion of these anonlalies. line in Figure 1.
20 Journal of Economic Literature, Vol. X X X V (March 1997)

estrrnation post-event
( I ] LsEv] ( ~vinclo~v]
---I---
T,, T, 0 T1. T,3
K'

F i g ~ ~ r1.e Time line for an event study.

I t is typical for the estimation window

and the event window not to overlap.
This design provides estimators for the
parameters of the normal return inodel
which are not influenced by the returns
around the event. Including tlle event where
window in the estimation of tlle normal
model parameters could lead to the
event returns having a large influence
on the norrnal return measure. In
this situation both the normal returns
and the abnorrnal returns would cap-
ture the event impact. This would be and
problematic because the inethodology
is built around the assumption that Ri, and R,,,, are the return in event pe-
the event irnpact is captured by the riod T for security i and the rnarket re-
abnormal returns. On occasion, the spectively. The use of the OLS estiina-
post event window data is included tors to nleasure abnorrnal returns and to
with the estimation window data to develop their statistical properties is ad-
estimate the norinal return inodel. dressed next. First, the properties of a
The goal of this approach is to increase given security are presented followed by
the robustness of the norrnal market consideration of the properties of abnor-
return rneasure to gradual changes mal returns aggregated across securities.
in its parameters. In Section 6 ex-
panding the null hypothesis to accoin- B . Statistical Properties of Abnorrl~al
inodate changes in the risk of a firm Returns
around the event is considered. In this case Given the market nlodel parameter
an estimation framework which uses the estimates, one can nleasure and analyze
event window returns will be required. the abnorrnal returns. Let ARiT, T = T 1 +
A. Estimation of the Marlcet Model 1, . . . , T,, be the sample of Lq abnornlal
returns for firrn i in the event window.
Under general conditions ordinary Using the market inodel to measure the
least squares (OLS) is a consistent esti- norinal return, tlle sample abnorinal re-
mation procedure for the market model turn is
parameters. Further, given the assump-
tions of Section 4, OLS is efficient. For
the it11 firm in event time, the OLS esti- The abnorinal return is the disturbance
mators of the rnarket model paranleters term of the market rnodel calculated on
for an estimation window of observations an out of sanlple basis. Under the null
are hypothesis, conditional on the event win-
 MacKinlay: Euent Studies i n Economics and Finance 21

dow inarket returns, the abnormal re- overall inferences for the event of inter-
turns will be jointly norinally distributed est. The aggregation is along two dimen-
with a zero conditional inean and condi- sions-through time and across securi-
tional variance o"ARiT) where ties. W e will first consider aggregation
through time for an individual security
and then will consider aggregation both
across securities and through tiine. The
concept of a cunlulative abnormal return
From (S), the conditional variance has
is necessary to acconlrnodate a multiple
two coinponents. One component is the
period event window. Define CARi(zl,z,)
disturbance variance oz, frorn ( 3 ) and a
as the sample cuinulative abnorinal re-
second component is additional variance
turn (CAR) froin 71 to 7 2 where
due to the salnpling error in al and Pi.
This sarnpling error, which is common
T1< zl 5 z2 5 T2.The CAR froin 7 , to T~ is
the sum of the included abnorrnal re-
for all the event window observations,
turns.
also leads to serial correlation of the
abnorrnal returns despite tlle fact that 7,

the true disturbances are independent CAR,(z,,z,) = AR,,. (10)

through tiine. As the length of the esti- 7 = 7,

nlation window L1 beconles large, the

second term approaches zero as the sam- Asyinptotically (as L1 increases) the vari-
pling error of the parameters vanishes. ance of CAR, is
The variance of the abnormal return will
be o:, and the abnorrnal return observa-
tions will become independent through
This large sarnple estimator of the vari-
tiine. I n practice, the estimation window
ance can be used for reasonable values of
can usually be chosen to be large enough
L1. However, for srnall values of L1 the
to rnake it reasonable to assunle that the
variance of the cuinulative abnorinal re-
contribution of the second conlponent to
turn should be adjusted for the effects of
the variance of the abnorrnal return is
the estimation error in the norinal model
zero.
parameters. This adjustment involves the
Under the null hypothesis, H o , that
second term of (S) and a further related
the event has no iinpact on tlle be-
adjustment for the serial covariance of
havior of returns (mean or variance)
the abnormal return.
the distributional properties of the
The distribution of the cumulative ab-
abnorrnal returns can be used to draw
norrnal return under H o is
inferences over any period within the
i I

event window. Under H o the distribu- - N(O,o?(z,,z,)).

CARi(z1,z2) (12)
tion of the sarnple abnorrnal return of a
Given the null distributions of tlle abnor-
given observation in the event window is
rnal return and the cunlulative abnornlal
AR,, - N(O,o"AR,,)). (9) return, tests of the null hypothesis can
be conducted.
Next (9) is built upon to consider the ag- However, tests with one event obser-
gregation of the abnormal returns. vation are not likely to be useful so it is
necessary to aggregate. The abnormal re-
C. Aggregation of Abnormal Returns
turn observations must be aggregated for
The abnornlal return observations the event window and across observa-
must be aggregated in order to draw tions of the event. For this aggregation,
 Journal of Econonzic Literature, Vol. XXXV (March 1997)

TABLE 1

Market Model
Good News No News Bad News
Event
Day AR CAR AR CAR AR CAR
-20 ,093 ,093
-19 -.I77 -.084
-18 ,088 ,004
-17 ,024 ,029
-16 -.018 ,011
-15 -.040 -.029
-14 ,038 ,008
-13 ,056 ,064
-12 ,065 ,129
-11 ,069 ,199
-10 ,028 ,227
-9 ,155 ,382
-
-8
-i
,057
-.010
,438
,428
-6 ,104 ,532
-5 ,085 ,616
-4 ,099 ,715
-3 ,117 ,832
-2 ,006 ,838
-1 ,164 1.001
0 ,965 1.966
1 .251 2.217
2 -.014 2.203
3 -.I64 2.039
4 -.014 2.024
5 ,135 2.160
-
6 -.052
,060
2.107
2.167
8 ,155 2.323
9 -.008 2.315
10 ,164 2.479
11 -.081 2.398
12 -.058 2.341
13 -.I65 2.176
14 -.081 2.095
15 -.007 2.088
16 ,065 2.153
17 ,081 2.234
18 ,172 2.406
19 -.043 2.363
20 ,013 2.377
 MacKinlay: Erjent Studies i n Econonzics and Finance 23

TABLE 1(Cont.)

Constant Mean Return Model

Good News No News Bad News
AR CAR AR CAR AR CAR
,105 .lo5 ,019 ,019 -.077 -.077
-.235 -.I29 -.048 -.029 -.I42 -.219
,069 -.060 -.086 -.I15 -.043 -.262
-.026 -.086 -.I40 -.255 -.057 -.319
-.086 -.I72 ,039 -.216 -.075 -.394
-.I83 -.355 ,099 -.I17 -.037 -.431
-.020 -.375 -. 150 -.266 -.lo1 -.532
-.025
,101
-.399
-.298
-.I91
,133
-.458
-.325
-.069
-.lo6 --
-.GO1
-.I01
,126 -.I72 ,006 -.319 -.I69 -.876
,134 -.038 ,103 -.216 -.009 -.885
,210 ,172 ,022 -.I94 ,011 -.874
,106 ,278 ,163 .-031 ,135 -.738
-.002 ,277 ,009 -.022 -.027 -.765
,011 ,288 -.029 -.051 ,030 -.735
,061 ,349 -.068 -.I20 ,320 -.415
,031 ,379 ,089 -.031 -.205 -.620
,067 ,447 ,013 -.018 ,085 -.536
,010 ,456 ,311 ,294 -.256 -.791
,198 ,654 -.I70 ,124 -.227 -1.018
1.034 1.688 -.I64 -.040 -.643 -1.661
,357 2.045 -.I70 -.210 -.212 -1.873
-.013 2.033 ,054 -.I56 ,078 -1.795
,088 1.944 -.I21 -.277 ,146: -1.648
,041 1.985 ,023 -.253 ,149 -1.499
,248 2.233 -.003 -256 ,286 -1.214
-.035 2.198 -.319 -.575 ,070 -1.143
,017 2.215 -. 112 -.687 ,102 -1.041
,112 2.326 -.I87 -.874 ,056 -.986
-.052 2.274 -.057 -.931 -.071 -1.056
,147 2.421 ,203 -.728 ,267 -.789
-.013 2.407 ,045 -.683 ,006 -.783
-.054 2.354 ,299 -.384 ,017 -.766
-.246 2.107 -.067 -.451 ,114 -.652
-.011 2.096 -.024 -.475 ,089 -.564
-.027 2.068 -.059 -.534 -.022 -.585
,103 2.171 -.046 -.580 -.084 -.670
,066 2.237 -.098 -.677 -.054 -.724
,110 2.347 ,021 -.656 -.071 -.795
-.055 2.292 ,088 -.568 ,026 -.769
,019 2.311 ,013 -554 -.I15 -.884

Abnormal retulns for an event study of the information content of earnings announcements. The sample consists of
a total of 600 quarterly announcements for the 30 companies in the Dow Jones Industrial Index for the five year
period Januay 1989 to December 1993. Two models are considered for the normal returns, the market model using
the CRSP value-weighted index and the constant return model. The announcements are categorized into three
groups, good news, no news, and bad news. AR is the sample average abnormal return for the specified day in event
time and CAR is the sample average cumulative abnormal return for day -20 to the specified day Event time is days
relative to the announcement date.
24 Journal of Economic Literature, Vol. XXXV (March 1997)

it is assumed that there is not any clus-

tering. That is, there is not any overlap
in the event windows of the included se-
curities. The absence of any overlap and
the maintained distributional assumptions For the variance estimators the assump-
imply that the abnormal returns and the tion that the event windows of the N se-
cumulative abnorrnal returns will be in- curities do not overlap is used to set the
dependent across securities. Later infer- covariance terms to zero. Inferences
ences with clustering will be discussed. about the cunlulative abnormal returns
The individual securities' abnormal re- can be drawn using
turns can be aggregated using AR,, b i n (7)
for each event period, z = T I + 1, . . . , Te
Given N events, the sainple aggregated to test the null hypothesis that the ab-
abnorinal returns for period T is norrnal returns are zero. I n practice, be-
Y
cause ozi is unknown, an estimator rnust
1 be used to calculate the variance of the
AT,=-xAR,,
N abnorrnal returns as in (14). The usual
i=l
sanlple variance measure of o:, frorn the
and for large L1, its variance is rnarket model regression in the estiina-
- A' tion window is an appropriate choice.
Using this to calculate vaI(AxT) in (14),
Ho can be tested using
Using these estimates, the abnorinal re-
turns for any event period can be ana-
lyzed.
The average abnorinal returns can This distributional result is asyrnptotic
then be aggregated over the event win- with respect to the nurnber of securities
dow using the same approach as that N and the length of estimation window LI.
used to calculate the curnulative abnor- Modifications to the basic approach
mal return for each security i. For any presented above are possible. One corn-
interval in the event window inon modification is to standardize each
abnornlal return using an estinlator of its
standard deviation. For certain alterna-
tives, such standardization can lead to
inore powerful tests. James Pate11 (1976)
presents tests based on standardization
and Brown and Warner (1980, 1985)
x
72

V ~ ~ ( C ~ ( T= ~ , var
T ~(AT,).
7 = 5,
)) (16)
provide comparisons with the basic ap-
proach.
D. CAR:s for tlze Earnings
Observe that equivalently one can forrn Announcement Example
the CAR'S security by security and then
The information content of earnings
aggregate through tirne,
exanlple previously described illustrates
the use of sample abnorrnal residuals and
sainple cumulative abnornlal returns. Ta-
ble 1 presents the abnorinal returns av-
 MacKinlay: Euent Studies in Economics and Finance 25

Event Tinie
--+
Good Neu-s Firms ----+-- No h'eu-s Firms Bad News Films
Figure 2a. Plot of culnulative abnormal return for earning announcements from event day -20 to event
day 20. The abnormal return is calculated using the market model as the normal return measure.

eraged across the 600 event observations deed convey infornlation useful for the
(30 firms, 20 announcenlents per firm) valuation of firms. Focusing on the an-
as well as the aggregated cumulative ab- nouncelllent day (day 0 ) the sainple aver-
normal return for each of the three earn- age abnornlal return for the good news
ings news categories. Two norrnal return firm using the market nlodel is 0.965
models are considered; the market percent. Given the standard error of the
nlodel and for comparison, the constant one day good news average abnormal re-
rnean return model. Plots of the cumula- turn is 0.104 percent, the value of O1 is
tive abnormal returns are also included, 9.28 and the null hypothesis that the
with the CAR's from the market inodel event has no impact is strongly rejected.
in Figure 2a and the CAR's frorn the The story is the same for the bad news
constant rnean return nlodel in Figure firms. The event day sainple abnormal
2b. return is -0.679 percent, with a standard
The results of this example are largely error of 0.098 percent, leading to O1
consistent with the existing literature on equal to -6.93 and again strong evidence
the information content of earnings. The against the null hypothesis. As would be
evidence strongly supports the hypothe- expected, the abnornlal return of the no
sis that earnings announceinents do in- news firrns is srnall at -0.091 percent and
26 Journal of Economic Literature, Vol. XXXV (March 1997)

-
- 0 . 0 2 5 0 1
-21 -18 -15 -12 -9 -6 -3 0 3 G 9 12 15
Event Time
C;ood News Firllls No News Firms ----t- Had News Firms
Figure 2b. Plot of cuinulative abnormal return for earning announcements froin event day -20 to event
day 20. The abnormal return is calculated using the constant mean return model as the normal return

with a standard error of 0.098 percent categories. When measuring abnormal

is less than one standard error from zero. returns with the constant mean return
There is some evidence of the announce- model the standard errors increase from
ment effect on day one. The average 0.104 percent to 0.130 percent for good
abnormal return is 0.251 percent and news firms, from 0.098 percent to 0.124
-0.204 percent for the good news and percent for no news firms, and from
the bad news firms respectively. Both 0.098 percent to 0.131 percent for bad
these values are more than two standard news firms. These increases are to be ex-
errors from zero. The source of these pected when considering a sample of
day one effects is likely to be that some large firms such as those in the Dow In-
of the earnings announcements are made dex because these stocks tend to have an
on event day zero after the close of the important market component whose vari-
stock market. In these cases, the effects ability is eliminated using the market
will be captured in the return on day model.
one. The CAR plots show that to some ex-
The conclusions using the abnormal tent the market gradually learns about
returns from the constant return model the forthcoming announcement. The av-
are consistent with those from the mar- erage CAR of the good news firms
ket model. However, there is some loss gradually drifts up in days -20 to -1
of precision using the constant return and the average CAR of the bad news
model, as the variance of the average ab- firms gradually drifts down over this
normal return increases for all three period. In the days after the an-
 MacKinlay: Event Studies in Economics and Finance 27

nouncement the CAR is relatively stable approach, an alternative hypothesis

as would be expected, although there where some of the firms have positive
does tend to be a slight (but statis- abnormal returns and some of the firms
tically insignificant) increase with the have negative abnormal returns can be
bad news firms in days two through accommodated. However, in general
eight. the approach has two drawbacks-fre-
quently the test statistic will have
E . Inferences tuith Clustering
poor finite sample properties except in
The analysis aggregating abnormal re- special cases and often the test will
turns has assumed that the event win- have little power against economically
dows of the included securities do not reasonable alternatives. The multivariate
overlap in calendar time. This assump- framework and its analysis is similar
tion allows us to calculate the variance of to the analysis of multivariate tests
the aggregated sample cumulative abnor- of asset pricing models. MacKinlay
mal returns without concern about the (1987) provides analysis in that con-
covariances across securities because text.
they are zero. However, when the event
windows do overlap and the covariances
between the abnormal returns will not 6 . Modijiying the Null Hypothesis
be zero, the distributional results pre-
sented for the aggregated abnormal re- Thus far the focus has been on a single
turns are no longer applicable. Victor null hypothesis-that the given event has
Bernard (1987) discusses some of the no impact on the behavior of the returns.
problems related to clustering. With this null hypothesis either a mean
Clustering can be accommodated in effect or a variance effect will represent
two ways. The abnormal returns can be a violation. However, in some applica-
aggregated into a portfolio dated using tions one may be interested in testing for
event time and the security level analysis a mean effect. In these cases, it is neces-
of Section 5 can applied to the portfolio. sary to expand the null hypothesis to al-
This approach will allow for cross corre- low for changing (usually increasing)
lation of the abnormal returns. variances. To allow for changing variance
A second method to handle clustering as part of the null hypothesis, it is neces-
is to analyze the abnormal returns with- sary to eliminate the reliance on the
out aggregation. One can consider test- past returns to estimate the variance of
ing the null hypothesis of the event hav- the aggregated cumulative abnormal re-
ing no impact using unaggregated turns. This is accomplished by using the
security by security data. This approach cross section of cumulative abnormal re-
is applied most commonly when there is turns to form an estimator of the vari-
total clustering, that is, there is an event ance for testing the null hypothesis.
on the same day for a number of firms. Ekkehart Boehmer, Jim Musumeci, and
The basic approach is an application of Annette Poulsen (1991) discuss method-
a multivariate regression model with ology to accommodate changing vari-
dummy variables for the event date. This ance.
approach is developed in the papers of The cross sectional approach to esti-
Katherine Schipper and Rex Thompson mating the variance can be applied to
(1983, 1985) and Daniel Collins and the
--
average cumulative abnormal return
Warren Dent (1984). The advantage of (CAR (T,,T~)).Using the cross-section to
the approach is that, unlike the portfolio form an estimator of the variance gives
28 Journal of Economic Literature, Vol. XXXV (March 1997)

Consider a two-sided test of the null

hypothesis using the cumulative abnor-
mal return based statistic 8, from (20).
It is assumed that the abnormal returns
are uncorrelated across securities; thus
3'
the variance of CARis 1 / ~ " ~?(zl,zz)
i=l
For this estimator of the variance to be
and N is the sample size. Because the
consistent, the abnormal returns need to
be uncorrelated in the cross-section. An null distribution of 8, is standard normal,
absence of clustering is sufficient for this for a two sided test of size a , the null
requirement. Note that cross-sectional hypothesis will be rejected if 81 is in the
homoskedasticity is not required. Given critical region, that is,
this variance estimator, the null hypothe-
sis that the cumulative abnormal returns
are zero can then be tested using the
usual theory. where c(x) = $-'(x). $(.) is the standard
One may also be interested in the normal cumulative distribution function
question of the impact of an event on the (CDF).
risk of a firm. The relevant measure of Given the specification of the alterna-
risk must be defined before this question tive hypothesis H A and the distribution
can be addressed. One choice as a risk of for this alternative, the power of a
measure is the market model beta which test of size a can be tabulated using the
is consistent with the Capital Asset Pric- power function,
ing Model being appropriate. Given this
choice, the market model can be formu-
lated to allow the beta to change over
the event window and the stability of the
risk can be examined. Edward Kane and
Haluk Unal (1988) present an applica-
tion of this idea.

7 . Analysis of Pozoer The distribution of 81 under the alterna-

tive hypothesis considered below will be
An important consideration when set- normal. The mean will be equal to the
ting up an event study is the ability to true cumulative abnormal return divided
detect the presence of a non-zero abnor- by the standard deviation of CAR and
mal return. The inability to distinguish the variance will be equal to one.
between the null hypothesis and eco- To tabulate the power one must posit
nomically interesting alternatives would economically plausible scenarios. The al-
suggest the need for modification of the ternative hypotheses considered are
design. In this section the question of four levels of abnormal returns, 0.5
the likelihood of rejecting the null hy- percent, 1.0 percent, 1.5 percent, and
pothesis for a specified level of abnormal 2.0 percent and two levels of the aver-
return associated with an event is ad- age variance for the cumulative abnor-
dressed. Formally, the power of the test mal return of a given security over the
is evaluated. event period, 0.0004 and 0.0016. The
 MacKinlay: Event Studies i n Economics and Finance 29

TABLE 2

Abllorlnal Return Abnorlnal Return

Sample ,005 ,010 ,015 ,020 ,005 ,010 ,015 ,020
Size 0 = 0.02 0 = 0.04

Power of event study illethodology for test of the null hypothesis that the abnormal return is zero. The power is
reported for a two-sided test using 81 with a size of 5 percent. The sample size is the number of event observations
included the study and 0 is the square root of the average variance of the abnorlnal return across firms.

sample size, that is the number of securi- ues calculated using c ( a / 2 ) and c ( 1 -
ties for which the event occurs, is a / 2 ) are -1,96 and 1.96 respectively. Of
varied from one to 200. The power for course, in applications, the power of the
a test with a size of 5 percent is docu- test should be considered when selecting
mented. With a = 0.05, the critical val- the size.
30 Journal of Econo~nicLiterature, Vol. XXXV (March 1997)

N l u i ~ l ~of
e r Securities
Figure 3a. 1'on.c.r of everit study test statistic 0, to reject tlie lilill li~potliesisthat tlie al~riormalretlnn is zero, IT-Ire11the
sqilare root of die a\wage \.arialice of tlre n1)lioririal retl~rnacross firins 1s 2 percerit.

The power results are presented in Ta- are substantial when the abnormal
ble 2, and are plotted in Figures 3a and return is larger. For example, when the
3b. The results in the left panel of Table abnormal return is 2.0 percent the
2 and Figure 3a are for the case where power of a 5 percent test with 20 firms
the average variance is 0.0004. This cor- is almost 1.00 with a value of 0.99.
responds to a cuinulative abnormal re- The general results for a variance of
turn standard deviation of 2 percent and 0.0004 is that when the abnormal return
is an appropriate value for an event is larger than 1 percent the power is
which does not lead to increased vari- quite high even for small sample sizes.
ance and can be examined using a one- When the abnormal return is small a
day event window. In terms of having larger sainple size is necessary to achieve
high power this is the best case scenario. high power.
The results illustrate that when the ab- In the right panel of Table 2 and in
normal return is only 0.5 percent the Figure 3b the power results are pre-
power can be low. For example with a sented for the case where the average
sample size of 20 the power of a 5 variance of the cumulative abnormal re-
percent test is only 0.20. One needs a turn is 0.0016. This case corresponds
sample of over 60 firms before the roughly to either a multi-day event win-
power reaches 0.50. However, for a dow or to a one-day event window with
given sample size, increases in power tlle event leading to increased variance
 MacKinlay: Euent Studies i n Econo~nicsand Finance 31

Figure 3b. Poxver of event stildy test statistic 8, to reject the nil11 li~potliesist1i;tt tlie a11norin;tl retilni is zero, wlien
the scp~awroot of tlie average varialice of the al)norii~;~l
return across Arms is 3 percent.

which is accoininodated as part of the considered analytically for the given dis-
null hypothesis. When the average vari- tributional assumptions. If the distri-
ance of the CAR is increased from butional assumptions are inappropriate
0.0004 to 0.0016 there is a dramatic then the results may differ. However,
power decline for a 5 percent test. When Brown and Warner (1985) consider this
the CAR is 0 . 5 percent the power is only possible difference and find that the ana-
0.09 with 20 firins and is only 0.42 with a lytical computations and the empirical
sample of 200 firms. This magnitude of power are very close.
abnormal return is difficult to detect I t is difficult to make general conclu-
with the larger variance. In contrast, sions concerning the adequacy of the
when the CAR is as large as 1.5 percent ability of event study methodology to de-
or 2.0 percent the 5 percent test is still tect non-zero abnormal returns. When
has reasonable power. For example, conducting an event study it is best
when the abnormal return is 1.5 percent to evaluate the power given the parame-
and there is a sample size of 30 the ters and objectives of the study. If the
power is 0.54. Generally if the abnormal power seems sufficient then one can
return is large one will have little diffi- proceed, otherwise one should search
culty rejecting the null hypothesis of no for ways of increasing the power. This
abnormal return. can be done by increasing the sample
In the preceding analysis the power is size, shortening the event window, or by
32 Journal of Econonzic Literature, V o l . XXXV ( M a r c h 1997)

developing more specific predictions to Charles Corrado (1989) proposes a non-

test. parametric rank test for abnormal per-
formance in event studies. A brief de-
8. Nonpnmnzetric Tests scription of his test of no abnormal
return for event day zero follows. The
The methods discussed to this point
framework can be easily altered for more
are parainetric in nature, in that specific
general tests.
assumptions have been made about
Drawing on notation previously intro-
the distribution of abnormal returns.
duced, consider a sample of L2 abnormal
Alternative approaches are available
returns for each of N securities. To im-
which are nonparametric in nature.
plement the rank test, for each security
These approaches are free of specific
it is necessary to rank tlie abnormal re-
assump- tions concerning the distri-
turns from one to L2. Define K,, as
bution of returns. Common nonpararnet-
the rank of the abnormal return of
ric tests for event studies are the sign
security i for event time period T. Re-
test and the rank test. These tests are dis-
call, z ranges from T I + 1 to T2 and T = 0
cussed next.
is the event day. Tlie rank test uses the
Tlie sign test, which is based on the
fact that the expected rank of the event
sign of tlie abnormal return, requires
day is ( L 2 + 1)/2 under the null hypothe-
that the abnormal returns (or more gen-
sis. The test statistic for the null hy-
erally cumulative abnormal returns) are
pothesis of no abnormal return on event
independent across securities and that
day zero is
the expected proportion of positive ab-
normal returns under the null hypothesis
is 0.5. The basis of the test is that, under
the null hypothesis, it is equally probable
that the CAR will be positive or nega- where
tive. If, for example, the null hypothesis
is that there is a positive abnormal re-
turn associated with a given event, the
null hypothesis is Ho:p 1 0.5 and the al-
Tests of the null hypothesis can be im-
ternative is Hll:p > 0.5 where p =
plemented using the result that tlie as-
pr[CAR, 2 0.01. To calculate the test sta-
ymptotic null distribution of 0:i is stan-
tistic we need the number of cases where
dard normal. Corrado (1989) includes
the abnormal return is positive, N+, and
further discussion of details of this test.
tlle total number of cases, N . Letting 02
Typically, these nonparametric tests
be the test statistic,
are not used in isolation but in conjunc-
tion with the parametric counterparts.
Inclusion of the nonparai~letrictests pro-
vides a check of the robustness of con-
This distributional result is asymptotic. clusions based on parametric tests. Such
For a test of size (1- a ) , Ho is rejected if a check can be worthwhile as illustrated
02 > @-'(a). by t l ~ ework of Cynthia Campbell and
A weakness of tlie sign test is that it Charles Wasley (1993). They find that
may not be well specified if the distri- for NASDAQ stocks daily returns the
bution of abnormal returns is skewed as nonparametric rank test provides more
can be the case with daily data. In re- reliable inferences than do the standard
sponse to this possible shortcoming, parametric tests.
 iMacKinlay: Event Studies i n Economics a n d Finance 33

9. Cross-Sectional Moclels turn in tlle eleven inonths prior to tlle

announcement month. They rind that the
Theoretical insights can result froin
magnitude of the (negative) abnorinal re-
examining the association between the
turn associated with the announcement
magnitude of the abnorlnal return and
of equity offerings is related to both
characteristics specific to the event ob-
these variables. Larger pre-event cumu-
servation. Often such an exercise can be
lative abnorinal returns are associated
llelpful when multiple hypotheses exist
with less negative abnormal returns and
for the source of the abnorlnal return. A
larger offerings are associated wit11 inore
cross-sectional regression model is an
negative abnormal returns. These find-
appropriate tool to investigate this asso-
ings are consistent with theoretical pre-
ciation. The basic approach is to run a
dictions wl~ichthey discuss.
cross-sectional regression of the abnor-
Issues concerning the interpretation of
mal returns on the cl~aracteristicsof in-
the results can arise with the cross-sec-
terest.
tional regression approach. In many
Given a sample of N abnormal return
situations, the event window abnormal
observations and M characteristics, the
return will be related to firm characteris-
regression inodel is:
tics not only tl~roughthe valuation ef-
AR, = 60+ 6,ri,+ . . . + + 6nflaf,+ TJ (26) fects of the event but also through a rela-
tion between the firin characteristics and
E(rl,) = 0 (27) the extent to whicl~the event is antici-
where AR, is the j t l l abnorinal return ob- pated. This can happen when investors
servation, x,,+rn = 1, . . . , M, are M char- rationally use the firin characteristics
acteristics for the j t l l observation and 5 is to forecast the likelil~oodof the event
the zero mean disturbance term that is occurring. In these cases, a linear rela-
uncorrelated with the x's. 6,,,, 1n = 0, . . . , tion between the valuation effect of the
M are the regression coefficients. Tlle event and the firm cllaracteristic can be
regression model can be estimated using hidden. Paul Malatesta and Tholnpson
OLS. Assuming the qj's are cross-sec- (1985) and William Lanen and Thomp-
tionally uncorrelated and homoskedastic, son (1988) provide examples of this situ-
inferences can be conducted using the ation.
usual OLS standard errors. Alternatively, Technically, with the relation between
witllout assuming homoskedasticity, llet- the firm cl~aracteristicsand the degree
eroskedasticity-consistent t-statistics us- of anticipation of the event introduces a
ing standard errors can be derived using selection bias. The assumption that the
the approach of Halbert White (1980). regression residual is uncorrelated with
Tlle use of heteroskedasticity-consistent the regressors breaks down and the OLS
standard errors is advisable because estimators are inconsistent. Consistent
there is no reason to expect the residuals estimators can be derived by explicitly
of (26) to be homoskedastic. incorporating the selection bias. Sankar-
Paul Asquit11 and David Mullins sllan Acllarya (1988) and B. Espen
(1986) provide an example of this cross- Eckbo, Vojislav Maksimovic, and Josepll
sectional approach. Tlle two day cumula- Williams (1990) provide examples of this
tive abnormal return for the announce- approach. N. R. Prabhala (1995) pro-
ment of an equity offering is regressed vides a good discussion of this problem
on the size of the offering as a percent- and the possible solutions. He argues
age of the value of the total equity of the that, despite an incorrect specification,
firm and on the cumulative abnormal re- under weak conditions, the OLS ap-
34 Journal of Econonzic Literature, Vol. XXXV (March 1997)

, , , I t , , , , , , 1 8 8 t , , , I , , , , , , , , , , , , , , , ,
Oi
0 10 20 :30 40 50 60 70 80 90 100 110 120 1:30 140 150 160 170 180 190
N11niber of Securities
Figure 4. Po\ver of event sttlrly test statistic to reject thr 111111 I ~ y ~ ~ o t h rthat
s i s thr al~normal1-rt1u.n1s zrro, tor
s , tlie sqllai-e root of tlie aver'lge \'ariance of tlie al~noriiialrettim across fii-111s is 4
differrllt sainpling ~ n t r i ~ a l\vhe11
percrnt for tlie daily interval. Sizr of test is 5 prrcent.

proach can be used for inferences and more frequent sampling arises. To ad-
that the t-statistics can be interpreted as dress this question one needs to consider
lower bounds on the true significance the power gains from shorter intervals. A
level of the estimates. comparison of daily versus monthly data
is provided in Figure 4. The power of
10. Other Isszles the test of no event effect is plotted
against the alternative of an abnormal re-
A number of further issues often arise turn of one percent for 1 to 200 securi-
when conducting an event study. These ties. As one would expect given the
issues include the role of the sampling analysis of Section 7, the decrease in
interval, event date uncertainty, robust- power going from a daily interval to a
ness, and some additional biases. inonthly interval is severe. For example,
A. Role of Sanzpling Interval with 50 securities the power for a 5 per-
cent test using daily data is 0.94, whereas
Stock return data is available at differ- the power using weekly and nlonthly
ent sampling intervals, wit11 daily and data is only 0.35 and 0.12 respectively.
niontl~lyintervals being the most com- The clear message is that there is a sub-
mon. Given the availability of various in- stantial payoff in terms of increased
tervals, the question of the gains of using power from reducing the sampling inter-
 MacKinlay: Ezjent Studies i n Economics a n d Finance 35

val. Dale Morse (1984) presents detailed informal procedure works well and there
analysis of the choice of daily versus is little to gain from the more elaborate
monthly data and draws the saine conclu- estimation framework.
sion.
A sampling interval of one day is not
the shortest interval possible. With the The statistical analysis of Sections 4, 5,
increased availability of transaction data, and 6 is based on assuinption that re-
recent studies have used observation in- turns are jointly normal and temporally
tervals of duration shorter than one day. independently and identically distri-
However, the net benefit of intervals less buted. I n this section, discussion of the
than one day is unclear as soine compli- robustness of the results to departures
cations are introduced. Discussion of us- from this assuinption is presented. The
ing transaction data for event studies is norlnality assuinption is important for
included in the work of Michael Barclay the exact finite sample results to hold.
and Robert Litzenberger (1988). Without assuming normality, all results
would be asymptotic. However, this is
B . Inferences w i t h Event-Date

generally not a problem for event studies

~&er-taint~

because for the test statistics, conver-

Thus far it is assumed that the event gence to the asymptotic distributions is
date can be identified with certainty. rather quick. Brown and Warner (1985)
However, in some studies it may be diffi- provide discussion of this issue.
cult to identify the exact date. A com-
D . O t h e r Possible Biases
mon example is when collecting event
dates from financial publications such as A nuinber of possible biases can arise
the W a l l Street Journal. When the event in the context of conducting an event
announcement appears in the paper one study. Nonsynchronous trading can in-
can not be certain if the inarket was in- troduce a bias. The nontrading or non-
forined prior to the close of the market synchronous trading effect arises when
the prior trading day. If this is the case prices, are taken to be recorded at time
then the prior day is the event day, if not intervals of one length when in fact they
then the current day is the event day are recorded at time intervals of other
The usual method of handling this prob- possibly irregular lengtl~s.For example,
lein is to expand the event window to the daily prices of securities usually em-
two days-day 0 and day + l . While there ployed in event studies are generally
is a cost to expanding the event window, "closing" prices, prices at which the last
the results in Section 6 indicated that transaction in each of those securities oc-
the power properties of two day event curred during the trading day. These
windows are still good suggesting that closing prices generally do not occur at
the costs are worth bearing rather than the saine time each day, but by calling
to take the risk of missing the event. tlleln "daily" prices, one is implicitly and
Clifford Ball and Walter Torous (1988) incorrectly assuming that they are
have investigated the issue They de- equally spaced at &-hour intervals. This
velop a maximum likelillood estiination nontrading effect induces biases in the
procedure which accommodates event lnoinents and co-moments of returns.
date uncertainty and examine results of The influence of the nontrading effect
their explicit procedure versus the infor- on the variances and covariances of indi-
mal procedure of expanding the event vidual stocks and portfolios naturally
window. The results indicates that the feeds into a bias for the inarket model
36 Tournnl of Econonzic Literature, Vol. XXXV (Marclz 1997)

beta. Myron Scholes and Williains (1977) have been in the area of corporate fi-
present a consistent estimator of beta in nance. Event studies dominate the em-
the presence of nontrading based on the pirical research in this area. Important
assuinption that the true return process exainples include the wealth effects of
is uncorrelated through time. They also mergers and acquisitions and the price
present soine empirical evidence which effects of financing decisions by firins
shows the nontrading-adjusted beta esti- Studies of these events typically focus on
mates of thinly traded securities to be the abnormal return around the date of
approximately 10 to 20 percent larger first announcement.
than the unadjusted estimates. However, In the 1960s there was a paucity of
for actively traded securities, the adjust- empirical evidence on the wealth effects
ments are generally slnall and unimpor- of mergers and acquisitions For exam-
tant. ple, Henry Manne (1965) discusses the
Prem Jain (1986) considers the influ- various arguments for and against merg-
ence of thin trading on the distribution ers. At that time the debate centered on
of the abnormal returns from the inarket the extent to which mergers should be
model with the beta estimated using the regulated in order to foster competition
Scholes-Williams approach. When coin- in the product markets Manne argued
paring the distribution of these abnormal that inergers represent a natural out-
returns to the distribution of the abnor- come in an efficiently operating market
lnal returns using the usual OLS betas for corporate control and consequently
finds that the differences are minimal. provide protection for shareholders. H e
This suggests that in general the adjust- downplayed the importance of the argu-
ments for thin trading are not important. inent that mergers reduce competition.
The methodology used to compute the At the conclusion of his article Manne
cuinulative abnormal returns can induce suggested that the two competing hy-
an upward bias. Tlle bias arises from the potheses for mergers could be separated
observation by observation rebalancing by studying the price effects of the in-
to equal weights implicit in the calcula- volved corporations. H e llypothesized
tion of the aggregate cuinulative abnor- that, if inergers created inarket power,
inal return combined with the use of one would observe price increases for
transaction prices which can represent both the target and acquirer. In contrast,
both the bid and the offer side of the if the merger represented the acquiring
market. Marshall Blulne and Robert corporation paying for control of the tar-
Stainbaugll (1983) analyze this bias and get, one would observe a price increase
sllow that it can be important for studies for the target only and not for the ac-
using low market capitalization firms quirer. However, Manne concludes, in
wl~ichhave, in percentage terms, wide reference to the price effects of mergers,
bid offer spreads. I n these cases the bias that "no data are presently available on
can be elilninated by considering cumu- this subject "
lative abnormal returns wl~ichrepresent Since that time an enormous body of
buy and hold strategies. empirical evidence on mergers and ac-
quisitions has developed which is domi-
11. Concluding Discussion nated by the use of event studies. The
general result is that, given a successful
In closing, exainples of event study takeover, the abnormal returns of the
successes and limitations are presented. targets are large and positive and the ab-
Perhaps the most successful applications normal returns of the acquirer are close
 MacKinlay: Event Studies in Econonzics and Finance 37

to zero. Gregg Jarrell and Poulsen (1989) -3.56 percent. In contrast, when firlns
document that tlle average abnormal re- decide to use straight debt financing, the
turn for target sl~arel~olders exceeds 20 average abnormal return is closer to
percent for a sample of 663 successful zero. Mikkelson and Partcll (1986) find
takeovers from 1960 to 1985. In contrast the average abnormal return for debt is-
the abnorinal returns for acquirers is sues to be -0.23 percent for a sainple of
close to zero. For the same sample, Jar- 171 issues. Findings sucll as these pro-
re11 and Poulsen find an average abnor- vide the fuel for the development of new
mal return of 1.14 percent for acquirers. theories. For example, in this case, the
In the 1980s they find the average abnor- findings inotivate the pecking order the-
inal return is negative at -1.10 percent. ory of capital structure developed by Ste-
Eckbo (1983) explicitly addresses the wart Myers and Nicholas Majluf (1984).
role of increased market power in ex- A major success related to those in the
plaining merger related abnormal re- corporate finance area is the implicit ac-
turns. H e separates mergers of compet- ceptance of event study lnetl~odologyby
ing firms from other mergers and finds tlle U.S. Supreme Court for determining
no evidence that tlle wealtll effects for materiality in insider trading cases and
competing firins are different. Further, for determining appropriate disgorge-
lle finds no evidence that rivals of firms ment amounts in cases of fraud. This im-
merging horizontally experience negative plicit acceptance in the 1988 Basic, In-
abnormal returns. From this lle con- corporated v. Levinson case and its
cludes that reduced coinpetition in the importance for securities law is discussed
product market is not an iinportant ex- in Mitchell and Netter (1994).
planation for merger gains. This leaves There have also been less successful
competition for corporate control a more applications. An iinportant cllaracteristic
likely explanation. Much additional em- of a successful event study is the ability
pirical work in tlle area of mergers and to identify precisely the date of the
acquisitions has been conducted. Mi- event. In cases where the event date is
chael Jensen and Richard Ruback (1983) difficult to identify or the event date is
and Jarrell, Jaines Brickley, and Netter partially anticipated, studies have been
(1988) provide detailed surveys of this less useful. For example, tlle wealth ef-
work. fects of regulatory changes for affected
A number of robust results have been entities can be difficult to detect using
developed froin event studies of financ- event study metl~odology.Tlle problem
ing decisions by corporations. When a is that regulatory changes are often de-
corporation announces that it will raise bated in tlle political arena over time and
capital in external markets there is, on any accoinpanying wealth effects gener-
average, a negative abnormal return. The ally will gradually be incorporated into
magnitude of the abnorinal return de- the value of a corporation as tlle prob-
pends on the source of external financ- ability of tlle change being adopted in-
ing. Asquith and Mullins (1986) find for creases.
a sample of 266 firlns announcing an eq- Larry Dann and Christopher Jaines
uity issue in the period 1963 to 1981 tlle (1982) discuss this issue in the context of
two day average abnormal return is -2.7 the impact of deposit interest rate ceil-
percent and on a sample of 80 firlns for ings for thrift institutions. In their study
the period 1972 to 1982 Wayne Mikkel- of changes in rate ceilings, they decide
son and Megan Partch (1986) find tlle not to consider a change in 1973 because
two day average abnormal return is it was due to legislative action. Schipper
38 Journal of Economic Literature, Vol. XXXV (iMarch 1997)

and Thoinpson (1983, 1985) also encoun- Based Accounting Research," J. Ace. Res., 1987,
25(1), p p 1-48.
ter this problem in a study of merger BLUME, MARSHALL E. A X D S T A ~ I B ~ I U G H ,
related regulations. They attempt to ROBERTF. "Biases in Computed Returns: An
circuinvent the problem of regulatory Application to the Size Effect," J . Finan. Econ. ,
Nov. 1983, 12(3),pp. 387-404.
changes being anticipated by identify- BOEHMER,EKKEHART;M u s u ~ ~ E ( : IJIM , C I I ~
ing dates when the probability of a POULSEN, ANXETTEB. "Event-Study Method-
regulatory change being passed changes. ology under Conditions of Event-Induced Vari-
ance," J . Finan. Econ., Dec. 1991, 30(2), pp.
However, they find largely insignificant 253-72.
results leaving open the possibility the BROII:~,STEPHENJ. tiKD WARXER,JEROLD B.

of absence of distinct event dates as "Measuring Security Price Performance," J. Fi-

nun. Econ., Sept. 1980, 8(3), 205-58.

the explanation of the lack of wealth ef- -- . "Usin Daily Stock Returns: The Case of

fects. Event studes,', J. Finan. Econ., Mar. 1985,

Much has been learned from the body 14(1),p p 3-31.

BROIVX, STEPHEX A N D WEIXSTEIN,MARK I.

of research based on the use of event "Derived Factors in Event Studies," J . Finan.

study methodology. In a general context, Econ., Sept. 1985, 14(3),pp. 491-95.

event studies have shown that, as would CAMPBELL, CYXTHIA J . AX11 WASLEY,CHARLES
E . "Measuring Security Price Performance Us-
be expected in a rational marketplace, ing Daily NASDAQ Returns," J. Finnn. Econ.,
prices do respond to new information. As Feb. 1993, 33(1), pp. 73-92.
one inoves forward, it is expected that COLLIKS, DtiKIEL W, A N D DEXT,WARREN T. "A
Comparison of Alternative Testing Methodolo-
event studies will continue to be a valu- gies Used In Capital Market Research," J. Ace.
able and widely used tool in economics Res., Spring 1984, 22(1), pp. 48-84.
and finance. CORRtIDo, CHARLES."A Nonparametric Test for
Abnormal Security-Price Performance in Event
Studies," J . Finan. Econ., Aug. 1989, 23(2), pp.
385-95.
ACHARYA, SAXKARSHAN. "A Generalized Econo- Dtixr\;, LARRYY. A X D Jr~b1!3s,CHRISTOPHER M.
metric Model and Tests of a Signalling Hy- "An Analysis of the Impact of Deposit Rate
pothesis with Two Discrete Signals," J. Finance, Ceilings on the Market Values of Thrift Institu-
June 1988, 43(2), pp. 413-29. tions," J. Finance, Dec. 1982, 37(5), pp. 1259-75.
ASHLEY,JOHXW. "Stock Prices and Changes in DOLLEY,JAMESCLAY."Characteristics and Proce-
Earnings and Dividends: Some Empirical Re- dure of Common Stock Split-Ups," Harvard
sults," J. Polit. Econ., Feb. 1962, 70(1), pp. 82- Bus. Rev., Apr. 1 9 3 3 , l l , pp. 316-26.
85. EC:KHO,B. ESPEN."Horizontal Mergers, Collu-
ASQUITH,PAUL A N D MULLINS,D ~ i v I n ."Equity sion, and Stockholder Wealth," J. Finan. Econ.,
Issues and Offering Dilution," J . Finnn. Econ., Apr. 1983, 11(1-4), pp. 241-73.
Jan./Feb. 1986, 15(1/2),pp. 61-89. E c ~ u o ,B. ESPEN;MAKSI~~OI'IC, VOJISLAI' AND

BALL, CLIFFORDA. A N D TOROUS,WALTERN. WILLIAMS,JOSEPH."Consistent Estimation of

"Investigating Security-Price Perforn~ance in Cross-Sectional Models in Event Studies," Rev.

the Presence of Event-Date Uncertainty," J . Fi- Financial Stud., 1990, 3(3), p p 343:65.

nan. Econ., Oct. 1988, 22(1), pp. 123-53. Ftibi:i, E U G E N EF . ET A L . "Tle Adjustment of
BALL, RAYA N D BROII'N, PHILIP. "An Empirical Stock Prices to New Information," Int. Econ.
Evaluation of Accounting Income Numbers," J. Rev., Feb. 1969, 10(1),pp. 1-21.
Ace. Res. , Autumn 1968, 6(2), pp. 159-78. FAMA,EUC:ENEF. A N D FRENCH,KENNETHR.
BARC:LAY,MICHAEL J. A N D LITZENHERGER, "Multifactor Explanations of Asset Pricing
ROBERTH. "Announcement Effects of New Anomalies," J . Finance, Mar. 1996, 51(1), pp.
Equity Issues and the Use of Intraday Price 55-84.
Data," J. Finnn. Econ., May 1988, 21(1), pp. JAIN,PREM."Analyses of the Distribution of Secu-
71-99. rity Market Model Prediction Errors for Daily
BARKER, C. AUSTIN."Effective Stock Splits," Har- ~ e t u r n sD ata," J . Acc. Res., Spring 1986, 24(lj,
vard Bus. Rev., Jan./Feb. 1956, 34(1), pp. 101- DD. 76-96.
06. J A ~ ~ E LGREGGL, A,; BRICKLEY, JAMESA. A N D
-- . "Stock Splits in a Bull Market," Hnrvnrd NETTER,JEFFRYM. "The Market for Corpo-
Bus. Rev., May/June 1957, 35(3), pp. 72-79. rate Control: The Empirical Evidence Since
-- . "Evaluation of Stock Dividends," Harvard 1980,"J . Econ. Perspectives, Winter 1988, 2(1),
Bus. Rev., July/Aug. 1958, 36(4), pp. 99-114. pp. 49-68.
BERNARD, VICTOR L. "Cross-Sectional De en JARRELL, GREGGA N D POULSEN, ANNETTE."The
dence and Problems in Inference in Maset: Returns to Acquiring Firms in Tender Offers:
 MacKinlay: Event Studies i n Economics and Finance 39
Evidence from Three Decades," Financial Man- MYERS,STEII'ARTC. A N D MAJLUF,NICHOLAS S.
agement, Autumn 1989,18(3),pp. 12-19. "Corporate Financing and Investment Deci-
JENSEN,MICHAELC. AND RUBACK, R I C H I ~ RS.
D sions When Firms Have Information That In-
"The Market for Corporate Control: The Scien- vestors Do Not Have," J. Finan. Econ., June
tific Evidence," J. Finan. Econ., Apr. 1983, 1984, 13(2),pp. 187-221.
11(1-4), pp. 5-50. PATELL,JAMESM. "Corporate Forecasts of Earn-
KANE,EDIVARII J. A N D UNAL,HALUK."Change ings Per Share and Stock Price Behavior: Em-
in Market Assessments of Deposit-Institution pirical Tests," J . Ace. Res., Autumn 1976, 14(2),
Riskiness," J . Finan. Services Res., June 1988, p p 246-76.
1(3), p p 207-29. PRAUHALA, N. R. "Conditional Methods in Event
LANEN, WILLIAM N. A N D THOMPSON,REX. Studies and an Equilibrium Justification for Us-
"Stock Price Reactions as Surrogates for the ing Standard Event Study Procedures." Work-
Net Cash-Flow Effects of Corporate Policy De- ing Paper. Yale U., Sept. 1995.
cisions," J. Ace. Econ., Dec. 1988, 10(4), pp. RITTER,JAYR. "Long-Run Performance of Initial
311-34. Public Offerings," J. Finance, Mar. 1991, 46(1),
LINTNER,JOHN."The Valuation of Risky Assets pp. 3-27.
and the Selection of Risky Investments in Stock R&<s, STEPHENA. "The Arbitrage Theory of
Portfolios and Capital Budgets," Rev. Econ. Capital Asset Pricing," J. E c o n . - ~ h e o r ~b,e c .
Stat., Feb. 1965, 47(1), pp. 13-37. 1976, 1313). VD. 341-60.
MAC:KINLAY, A. CRAIG."On Multivariate Tests of SC:HIPPER,K~XHERINE AND THOMPSON, REX.
the CAPM," J. Finan. Econ., June 1987, 18(2), "The Impact of Merger-Related Regulations
p p 341-71. on the Shareholders of Acquiring Firms,"
MALATESTA,PAUL H . A N D THOMPSON,REX.
"Partially Anticipated Events: A Model of Stock -".
J. Acc. Res., Spring 1983, 21(1), pp. 184-
ZZl.

Price Reactions with an Application to Corpo- -- . "The Impact of Merger-Related Regula-

rate Acquisitions," J . Finan. Econ., June 1985, tions Using Exact Distributions of Test Statis-
14(2),p p 237-50. tics," J. Acc. Res., Spring 1985, 23(1), pp. 408-
MANNE,H E N R YG. "Mergers and the Market for 15.
Corporate Control," J. Polit. Econ., Apr. 1965, SC:HOLES,MYRONA N D WILLIAMS,JOSEPH T.
73(2), pp. 110-20. "Estimating Betas from Nonsynchronous Data,"
MC:QUEEN,GRANTA N D ROLEY,VANC:E."Stock J. Finan. Econ., Dec. 1977, 5(3), pp. 309-
Prices, News, and Business Conditions," Reu. 27.
Finan. Stud., 1993, 6(3), pp. 683-707. SCHIVERT,G. WILLIAM."Using Financial Data to
MIKKELSON, WAYNEH. AND PARTCH,MEC:AN. Measure Effects of Regulation," J. Law Econ.,
"Valuation Effects of Security Offerings and the Apr. 1981,24(1),pp. 121-58.
Issuance Process," J . Finan. Econ., Jan./Feb. SHARPE,WILLIAMF . "Capital Asset Prices: A
1986, 15(1/2),pp. 31-60. Theory of Market Equilibrium Under Condi-
MITCHELL,MARKL. A N D NETTER,JEFFRYM. tions of Risk," J. Finance, Sept. 1964, 19(3),pp.
"The Role of Financial Economics in Securities 425-42.
Fraud Cases: Applications at the Securities and -- . Portfolio theory and capital ntarkets. New
Exchange Commission,'' Business Lawyer, Feb. York: McGraw-Hill, 1970.
1994, 49(2),pp. 545-90. SHARPE,WILLIAMF.; ALEXANIIER, GORDONJ.
MORSE,DALE."An Econometric Analysis of the AN11 BAILEY,JEFFERY V. Investnzents. Fifth
Choice of Daily Versus Monthly Returns In Ed. Englewood Cliffs, NJ: Prentice-Hall,
Tests of Information Content," J. Acc. Res., 1995.
Autumn 1984,22(2),pp. 605-23. WHITE, HALBERT."A Heteroskedasticity-Consis-
MYERS,JOHN H , ANII BAKAY, ARCHIEJ . "Influ- tent Covariance Matrix Estimator and a Direct
ence of Stock Split-Ups on Market Price," Har- Test for Heteroskedasticity," Econonzetrica,
uard Bus. Rev., Mar. 1948, 26, pp. 251-55. May 980, 48(4), pp. 817-38.
http://www.jstor.org

LINKED CITATIONS
- Page 1 of 3 -

You have printed the following article:

Event Studies in Economics and Finance
A. Craig MacKinlay
Journal of Economic Literature, Vol. 35, No. 1. (Mar., 1997), pp. 13-39.
Stable URL:
http://links.jstor.org/sici?sici=0022-0515%28199703%2935%3A1%3C13%3AESIEAF%3E2.0.CO%3B2-W

This article references the following linked citations. If you are trying to access articles from an
off-campus location, you may be required to first logon via your library web site to access JSTOR. Please
visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes]

2
Multifactor Explanations of Asset Pricing Anomalies
Eugene F. Fama; Kenneth R. FrencH
The Journal of Finance, Vol. 51, No. 1. (Mar., 1996), pp. 55-84.
Stable URL:
http://links.jstor.org/sici?sici=0022-1082%28199603%2951%3A1%3C55%3AMEOAPA%3E2.0.CO%3B2-B

References

A Generalized Econometric Model and Tests of a Signalling Hypothesis with Two Discrete
Signals
Sankarshan Acharya
The Journal of Finance, Vol. 43, No. 2. (Jun., 1988), pp. 413-429.
Stable URL:
http://links.jstor.org/sici?sici=0022-1082%28198806%2943%3A2%3C413%3AAGEMAT%3E2.0.CO%3B2-M

Stock Prices and Changes in Earnings and Dividends: Some Empirical Results
John W. Ashley
The Journal of Political Economy, Vol. 70, No. 1. (Feb., 1962), pp. 82-85.
Stable URL:
http://links.jstor.org/sici?sici=0022-3808%28196202%2970%3A1%3C82%3ASPACIE%3E2.0.CO%3B2-M

NOTE: The reference numbering from the original has been maintained in this citation list.
http://www.jstor.org

LINKED CITATIONS
- Page 2 of 3 -

An Analysis of the Impact of Deposit Rate Ceilings on the Market Values of Thrift Institutions
Larry Y. Dann; Christopher M. James
The Journal of Finance, Vol. 37, No. 5. (Dec., 1982), pp. 1259-1275.
Stable URL:
http://links.jstor.org/sici?sici=0022-1082%28198212%2937%3A5%3C1259%3AAAOTIO%3E2.0.CO%3B2-M

Consistent Estimation of Cross-Sectional Models in Event Studies

B. Espen Eckbo; Vojislav Maksimovic; Joseph Williams
The Review of Financial Studies, Vol. 3, No. 3. (1990), pp. 343-365.
Stable URL:
http://links.jstor.org/sici?sici=0893-9454%281990%293%3A3%3C343%3ACEOCMI%3E2.0.CO%3B2-Z

The Adjustment of Stock Prices to New Information

Eugene F. Fama; Lawrence Fisher; Michael C. Jensen; Richard Roll
International Economic Review, Vol. 10, No. 1. (Feb., 1969), pp. 1-21.
Stable URL:
http://links.jstor.org/sici?sici=0020-6598%28196902%2910%3A1%3C1%3ATAOSPT%3E2.0.CO%3B2-P

Multifactor Explanations of Asset Pricing Anomalies

Eugene F. Fama; Kenneth R. FrencH
The Journal of Finance, Vol. 51, No. 1. (Mar., 1996), pp. 55-84.
Stable URL:
http://links.jstor.org/sici?sici=0022-1082%28199603%2951%3A1%3C55%3AMEOAPA%3E2.0.CO%3B2-B

The Valuation of Risk Assets and the Selection of Risky Investments in Stock Portfolios and
Capital Budgets
John Lintner
The Review of Economics and Statistics, Vol. 47, No. 1. (Feb., 1965), pp. 13-37.
Stable URL:
http://links.jstor.org/sici?sici=0034-6535%28196502%2947%3A1%3C13%3ATVORAA%3E2.0.CO%3B2-7

Mergers and the Market for Corporate Control

Henry G. Manne
The Journal of Political Economy, Vol. 73, No. 2. (Apr., 1965), pp. 110-120.
Stable URL:
http://links.jstor.org/sici?sici=0022-3808%28196504%2973%3A2%3C110%3AMATMFC%3E2.0.CO%3B2-3

NOTE: The reference numbering from the original has been maintained in this citation list.
http://www.jstor.org

LINKED CITATIONS
- Page 3 of 3 -

The Long-Run Performance of Initial Public Offerings

Jay R. Ritter
The Journal of Finance, Vol. 46, No. 1. (Mar., 1991), pp. 3-27.
Stable URL:
http://links.jstor.org/sici?sici=0022-1082%28199103%2946%3A1%3C3%3ATLPOIP%3E2.0.CO%3B2-9

Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk

William F. Sharpe
The Journal of Finance, Vol. 19, No. 3. (Sep., 1964), pp. 425-442.
Stable URL:
http://links.jstor.org/sici?sici=0022-1082%28196409%2919%3A3%3C425%3ACAPATO%3E2.0.CO%3B2-O

A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for

Heteroskedasticity
Halbert White
Econometrica, Vol. 48, No. 4. (May, 1980), pp. 817-838.
Stable URL:
http://links.jstor.org/sici?sici=0012-9682%28198005%2948%3A4%3C817%3AAHCMEA%3E2.0.CO%3B2-K

NOTE: The reference numbering from the original has been maintained in this citation list.