Matlab Econometrics Toolbox Documentation

download

https://ebookbell.com/product/matlab-econometrics-toolbox-
documentation-6741906

Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.

Matlab R2023a Econometrics Toolbox Users Guide Mathworks

https://ebookbell.com/product/matlab-r2023a-econometrics-toolbox-
users-guide-mathworks-49478912

Matlab Econometrics Toolbox Users Guide R2020a The Mathworks

https://ebookbell.com/product/matlab-econometrics-toolbox-users-
guide-r2020a-the-mathworks-11236174

Matlab A Practical Introduction To Programming And Problem Solving 6th

Edition Stormy Attaway

https://ebookbell.com/product/matlab-a-practical-introduction-to-
programming-and-problem-solving-6th-edition-stormy-attaway-46075418

Matlab For Engineering Applications 5e Ise 5thise William J Palm Iii

https://ebookbell.com/product/matlab-for-engineering-
applications-5e-ise-5thise-william-j-palm-iii-46562738
Matlab And Simulink Crash Course For Engineers 1st Ed 2022 Eklas
Hossain

https://ebookbell.com/product/matlab-and-simulink-crash-course-for-
engineers-1st-ed-2022-eklas-hossain-46608958

Matlab Medical Imaging Toolbox Users Guide The Mathworks

https://ebookbell.com/product/matlab-medical-imaging-toolbox-users-
guide-the-mathworks-46611260

Matlab For Engineering Applications 5th Edition William Palm

https://ebookbell.com/product/matlab-for-engineering-applications-5th-
edition-william-palm-46651690

Matlab For Engineering Berardino Dacunto

https://ebookbell.com/product/matlab-for-engineering-berardino-
dacunto-48710470

Matlab For Medical Physics Reallife Clinical Scenarios And Projects

Jidi Sun

https://ebookbell.com/product/matlab-for-medical-physics-reallife-
clinical-scenarios-and-projects-jidi-sun-49034592
Econometrics Toolbox™
User's Guide

R2016a
How to Contact MathWorks

Latest news: www.mathworks.com

Sales and services: www.mathworks.com/sales_and_services

User community: www.mathworks.com/matlabcentral

Technical support: www.mathworks.com/support/contact_us

Phone: 508-647-7000

The MathWorks, Inc.

3 Apple Hill Drive
Natick, MA 01760-2098
Econometrics Toolbox™ User's Guide
© COPYRIGHT 1999–2016 by The MathWorks, Inc.
The software described in this document is furnished under a license agreement. The software may be used
or copied only under the terms of the license agreement. No part of this manual may be photocopied or
reproduced in any form without prior written consent from The MathWorks, Inc.
FEDERAL ACQUISITION: This provision applies to all acquisitions of the Program and Documentation
by, for, or through the federal government of the United States. By accepting delivery of the Program
or Documentation, the government hereby agrees that this software or documentation qualifies as
commercial computer software or commercial computer software documentation as such terms are used
or defined in FAR 12.212, DFARS Part 227.72, and DFARS 252.227-7014. Accordingly, the terms and
conditions of this Agreement and only those rights specified in this Agreement, shall pertain to and
govern the use, modification, reproduction, release, performance, display, and disclosure of the Program
and Documentation by the federal government (or other entity acquiring for or through the federal
government) and shall supersede any conflicting contractual terms or conditions. If this License fails
to meet the government's needs or is inconsistent in any respect with federal procurement law, the
government agrees to return the Program and Documentation, unused, to The MathWorks, Inc.
Trademarks
MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See
www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand
names may be trademarks or registered trademarks of their respective holders.
Patents
MathWorks products are protected by one or more U.S. patents. Please see
www.mathworks.com/patents for more information.
Revision History
October 2008 Online only Version 1.0 (Release 2008b)
March 2009 Online only Revised for Version 1.1 (Release 2009a)
September 2009 Online only Revised for Version 1.2 (Release 2009b)
March 2010 Online only Revised for Version 1.3 (Release 2010a)
September 2010 Online only Revised for Version 1.4 (Release 2010b)
April 2011 Online only Revised for Version 2.0 (Release 2011a)
September 2011 Online only Revised for Version 2.0.1 (Release 2011b)
March 2012 Online only Revised for Version 2.1 (Release 2012a)
September 2012 Online only Revised for Version 2.2 (Release 2012b)
March 2013 Online only Revised for Version 2.3 (Release 2013a)
September 2013 Online only Revised for Version 2.4 (Release 2013b)
March 2014 Online Only Revised for Version 3.0 (Release 2014a)
October 2014 Online Only Revised for Version 3.1 (Release 2014b)
March 2015 Online Only Revised for Version 3.2 (Release 2015a)
September 2015 Online Only Revised for Version 3.3 (Release 2015b)
March 2016 Online Only Revised for Version 3.4 (Release 2016a)
 Contents

Getting Started
1
Econometrics Toolbox Product Description . . . . . . . . . . . . . 1-2
Key Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2

Econometric Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3

Model Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
Econometrics Toolbox Features . . . . . . . . . . . . . . . . . . . . . . . 1-3

Model Objects, Properties, and Methods . . . . . . . . . . . . . . . . 1-8

Model Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8
Model Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-9
Specify Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-11
Retrieve Model Properties . . . . . . . . . . . . . . . . . . . . . . . . . . 1-16
Modify Model Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-17
Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-18

Stochastic Process Characteristics . . . . . . . . . . . . . . . . . . . . 1-20

What Is a Stochastic Process? . . . . . . . . . . . . . . . . . . . . . . . 1-20
Stationary Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-21
Linear Time Series Model . . . . . . . . . . . . . . . . . . . . . . . . . . 1-22
Lag Operator Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-22
Characteristic Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-23
Unit Root Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-24

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-27

v
 Data Preprocessing
2
Data Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
Why Transform? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
Common Data Transformations . . . . . . . . . . . . . . . . . . . . . . 2-2

Trend-Stationary vs. Difference-Stationary Processes . . . . . 2-7

Nonstationary Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
Trend Stationary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-9
Difference Stationary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-9

Specify Lag Operator Polynomials . . . . . . . . . . . . . . . . . . . . 2-11

Lag Operator Polynomial of Coefficients . . . . . . . . . . . . . . . 2-11
Difference Lag Operator Polynomials . . . . . . . . . . . . . . . . . 2-14

Nonseasonal Differencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-18

Nonseasonal and Seasonal Differencing . . . . . . . . . . . . . . . . 2-23

Time Series Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . 2-28

Moving Average Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-31

Moving Average Trend Estimation . . . . . . . . . . . . . . . . . . . . 2-33

Parametric Trend Estimation . . . . . . . . . . . . . . . . . . . . . . . . 2-37

Hodrick-Prescott Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-45

Using the Hodrick-Prescott Filter to Reproduce Their

Original Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-46

Seasonal Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-51

What Is a Seasonal Filter? . . . . . . . . . . . . . . . . . . . . . . . . . 2-51
Stable Seasonal Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-51
Sn × m seasonal filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-52

Seasonal Adjustment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-54

What Is Seasonal Adjustment? . . . . . . . . . . . . . . . . . . . . . . 2-54
Deseasonalized Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-54
Seasonal Adjustment Process . . . . . . . . . . . . . . . . . . . . . . . 2-55

vi Contents
 Seasonal Adjustment Using a Stable Seasonal Filter . . . . . 2-57

Seasonal Adjustment Using S(n,m) Seasonal Filters . . . . . . 2-64

Model Selection
3
Box-Jenkins Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2

Box-Jenkins Model Selection . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4

Autocorrelation and Partial Autocorrelation . . . . . . . . . . . . 3-13

What Are Autocorrelation and Partial Autocorrelation? . . . . 3-13
Theoretical ACF and PACF . . . . . . . . . . . . . . . . . . . . . . . . . 3-13
Sample ACF and PACF . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-14

Ljung-Box Q-Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-16

Detect Autocorrelation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-18

Compute Sample ACF and PACF . . . . . . . . . . . . . . . . . . . . 3-18
Conduct the Ljung-Box Q-Test . . . . . . . . . . . . . . . . . . . . . . 3-21

Engle’s ARCH Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-25

Detect ARCH Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-28

Test Autocorrelation of Squared Residuals . . . . . . . . . . . . . 3-28
Conduct Engle's ARCH Test . . . . . . . . . . . . . . . . . . . . . . . . 3-31

Unit Root Nonstationarity . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-34

What Is a Unit Root Test? . . . . . . . . . . . . . . . . . . . . . . . . . 3-34
Modeling Unit Root Processes . . . . . . . . . . . . . . . . . . . . . . . 3-34
Available Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-39
Testing for Unit Roots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-40

Unit Root Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-44

Test Simulated Data for a Unit Root . . . . . . . . . . . . . . . . . . 3-44
Test Time Series Data for a Unit Root . . . . . . . . . . . . . . . . 3-50
Test Stock Data for a Random Walk . . . . . . . . . . . . . . . . . . 3-53

Assess Stationarity of a Time Series . . . . . . . . . . . . . . . . . . . 3-58

vii
 Test Multiple Time Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-62

Information Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-63

Model Comparison Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-65

Available Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-65
Likelihood Ratio Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-67
Lagrange Multiplier Test . . . . . . . . . . . . . . . . . . . . . . . . . . 3-67
Wald Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-68
Covariance Matrix Estimation . . . . . . . . . . . . . . . . . . . . . . 3-68

Conduct a Lagrange Multiplier Test . . . . . . . . . . . . . . . . . . . 3-70

Conduct a Wald Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-74

Compare GARCH Models Using Likelihood Ratio Test . . . . 3-77

Check Fit of Multiplicative ARIMA Model . . . . . . . . . . . . . . 3-81

Goodness of Fit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-88

Residual Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-90

Check Residuals for Normality . . . . . . . . . . . . . . . . . . . . . . 3-90
Check Residuals for Autocorrelation . . . . . . . . . . . . . . . . . . 3-90
Check Residuals for Conditional Heteroscedasticity . . . . . . 3-91

Check Predictive Performance . . . . . . . . . . . . . . . . . . . . . . . 3-92

Nonspherical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-94

What Are Nonspherical Models? . . . . . . . . . . . . . . . . . . . . . 3-94

Plot a Confidence Band Using HAC Estimates . . . . . . . . . . 3-95

Change the Bandwidth of a HAC Estimator . . . . . . . . . . . . 3-105

Check Model Assumptions for Chow Test . . . . . . . . . . . . . 3-112

Power of the Chow Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-123

viii Contents
 Time Series Regression Models
4
Time Series Regression Models . . . . . . . . . . . . . . . . . . . . . . . . 4-3

Regression Models with Time Series Errors . . . . . . . . . . . . . 4-6

What Are Regression Models with Time Series Errors? . . . . . 4-6
Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-7

Specify Regression Models with ARIMA Errors Using

regARIMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-10
Default Regression Model with ARIMA Errors Specifications 4-10
Specify regARIMA Models Using Name-Value Pair
Arguments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-12

Specify the Default Regression Model with ARIMA Errors 4-20

Modify regARIMA Model Properties . . . . . . . . . . . . . . . . . . . 4-22

Modify Properties Using Dot Notation . . . . . . . . . . . . . . . . . 4-22
Nonmodifiable Properties . . . . . . . . . . . . . . . . . . . . . . . . . . 4-25

Specify Regression Models with AR Errors . . . . . . . . . . . . . 4-29

Default Regression Model with AR Errors . . . . . . . . . . . . . . 4-29
AR Error Model Without an Intercept . . . . . . . . . . . . . . . . . 4-30
AR Error Model with Nonconsecutive Lags . . . . . . . . . . . . . 4-31
Known Parameter Values for a Regression Model with AR
Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-32
Regression Model with AR Errors and t Innovations . . . . . . 4-33

Specify Regression Models with MA Errors . . . . . . . . . . . . . 4-35

Default Regression Model with MA Errors . . . . . . . . . . . . . 4-35
MA Error Model Without an Intercept . . . . . . . . . . . . . . . . 4-36
MA Error Model with Nonconsecutive Lags . . . . . . . . . . . . . 4-37
Known Parameter Values for a Regression Model with MA
Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-38
Regression Model with MA Errors and t Innovations . . . . . . 4-39

Specify Regression Models with ARMA Errors . . . . . . . . . . 4-42

Default Regression Model with ARMA Errors . . . . . . . . . . . 4-42
ARMA Error Model Without an Intercept . . . . . . . . . . . . . . 4-43
ARMA Error Model with Nonconsecutive Lags . . . . . . . . . . 4-44

ix
 Known Parameter Values for a Regression Model with ARMA
Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-44
Regression Model with ARMA Errors and t Innovations . . . 4-45

Specify Regression Models with ARIMA Errors . . . . . . . . . 4-48

Default Regression Model with ARIMA Errors . . . . . . . . . . 4-48
ARIMA Error Model Without an Intercept . . . . . . . . . . . . . 4-49
ARIMA Error Model with Nonconsecutive Lags . . . . . . . . . . 4-50
Known Parameter Values for a Regression Model with ARIMA
Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-51
Regression Model with ARIMA Errors and t Innovations . . . 4-52

Specify Regression Models with SARIMA Errors . . . . . . . . 4-55

SARMA Error Model Without an Intercept . . . . . . . . . . . . . 4-55
Known Parameter Values for a Regression Model with SARIMA
Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-56
Regression Model with SARIMA Errors and t Innovations . . 4-57

Specify a Regression Model with SARIMA Errors . . . . . . . . 4-60

Specify the ARIMA Error Model Innovation Distribution . 4-69

About the Innovation Process . . . . . . . . . . . . . . . . . . . . . . . 4-69
Innovation Distribution Options . . . . . . . . . . . . . . . . . . . . . 4-70
Specify the Innovation Distribution . . . . . . . . . . . . . . . . . . . 4-71

Impulse Response for Regression Models with ARIMA

Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-75

Plot the Impulse Response of regARIMA Models . . . . . . . . . 4-77

Regression Model with AR Errors . . . . . . . . . . . . . . . . . . . . 4-77
Regression Model with MA Errors . . . . . . . . . . . . . . . . . . . 4-79
Regression Model with ARMA Errors . . . . . . . . . . . . . . . . . 4-80
Regression Model with ARIMA Errors . . . . . . . . . . . . . . . . 4-82

Maximum Likelihood Estimation of regARIMA Models . . . 4-86

Innovation Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-86
Loglikelihood Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-87

regARIMA Model Estimation Using Equality Constraints . 4-89

Presample Values for regARIMA Model Estimation . . . . . . 4-95

Initial Values for regARIMA Model Estimation . . . . . . . . . . 4-98

x Contents
Optimization Settings for regARIMA Model Estimation . . 4-100
Optimization Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-100
Constraints on Regression Models with ARIMA Errors . . . 4-103

Estimate a Regression Model with ARIMA Errors . . . . . . . 4-105

Estimate a Regression Model with Multiplicative ARIMA

Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-114

Select a Regression Model with ARIMA Errors . . . . . . . . . 4-123

Choose Lags for an ARMA Error Model . . . . . . . . . . . . . . . 4-125

Intercept Identifiability in Regression Models with ARIMA

Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-130
Intercept Identifiability . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-130
Intercept Identifiability Illustration . . . . . . . . . . . . . . . . . 4-132

Alternative ARIMA Model Representations . . . . . . . . . . . . 4-136

regARIMA to ARIMAX Model Conversion . . . . . . . . . . . . . 4-136
Illustrate regARIMA to ARIMAX Model Conversion . . . . . 4-137

Simulate Regression Models with ARMA Errors . . . . . . . . 4-145

Simulate an AR Error Model . . . . . . . . . . . . . . . . . . . . . . . 4-145
Simulate an MA Error Model . . . . . . . . . . . . . . . . . . . . . . 4-153
Simulate an ARMA Error Model . . . . . . . . . . . . . . . . . . . . 4-161

Simulate Regression Models with Nonstationary Errors . 4-171

Simulate a Regression Model with Nonstationary Errors . . 4-171
Simulate a Regression Model with Nonstationary Exponential
Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-175

Simulate Regression Models with Multiplicative Seasonal

Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-181
Simulate a Regression Model with Stationary Multiplicative
Seasonal Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-181
............................................. 4-184

Monte Carlo Simulation of Regression Models with ARIMA

Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-187
What Is Monte Carlo Simulation? . . . . . . . . . . . . . . . . . . . 4-187
Generate Monte Carlo Sample Paths . . . . . . . . . . . . . . . . . 4-187
Monte Carlo Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-189

xi
 Presample Data for regARIMA Model Simulation . . . . . . . 4-191

Transient Effects in regARIMA Model Simulations . . . . . . 4-192

What Are Transient Effects? . . . . . . . . . . . . . . . . . . . . . . . 4-192
Illustration of Transient Effects on Regression . . . . . . . . . 4-192

Forecast a Regression Model with ARIMA Errors . . . . . . . 4-202

Forecast a Regression Model with Multiplicative Seasonal

ARIMA Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-206

Verify Predictive Ability Robustness of a regARIMA Model 4-212

MMSE Forecasting Regression Models with ARIMA Errors 4-215

What Are MMSE Forecasts? . . . . . . . . . . . . . . . . . . . . . . . 4-215
How forecast Generates MMSE Forecasts . . . . . . . . . . . . . 4-216
Forecast Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-218

Monte Carlo Forecasting of regARIMA Models . . . . . . . . . 4-220

Monte Carlo Forecasts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-220
Advantage of Monte Carlo Forecasts . . . . . . . . . . . . . . . . . 4-220

Conditional Mean Models

5
Conditional Mean Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
Unconditional vs. Conditional Mean . . . . . . . . . . . . . . . . . . . 5-3
Static vs. Dynamic Conditional Mean Models . . . . . . . . . . . . 5-3
Conditional Mean Models for Stationary Processes . . . . . . . . 5-4

Specify Conditional Mean Models Using arima . . . . . . . . . . . 5-6

Default ARIMA Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6
Specify Nonseasonal Models Using Name-Value Pairs . . . . . . 5-8
Specify Multiplicative Models Using Name-Value Pairs . . . . 5-13

Autoregressive Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-18

AR(p) Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-18
Stationarity of the AR Model . . . . . . . . . . . . . . . . . . . . . . . 5-18

xii Contents
AR Model Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-21
Default AR Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-21
AR Model with No Constant Term . . . . . . . . . . . . . . . . . . . 5-22
AR Model with Nonconsecutive Lags . . . . . . . . . . . . . . . . . . 5-23
ARMA Model with Known Parameter Values . . . . . . . . . . . 5-24
AR Model with a t Innovation Distribution . . . . . . . . . . . . . 5-25

Moving Average Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-27

MA(q) Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-27
Invertibility of the MA Model . . . . . . . . . . . . . . . . . . . . . . . 5-27

MA Model Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-29

Default MA Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-29
MA Model with No Constant Term . . . . . . . . . . . . . . . . . . . 5-30
MA Model with Nonconsecutive Lags . . . . . . . . . . . . . . . . . 5-31
MA Model with Known Parameter Values . . . . . . . . . . . . . . 5-32
MA Model with a t Innovation Distribution . . . . . . . . . . . . . 5-32

Autoregressive Moving Average Model . . . . . . . . . . . . . . . . . 5-34

ARMA(p,q) Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-34
Stationarity and Invertibility of the ARMA Model . . . . . . . . 5-35

ARMA Model Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . 5-37

Default ARMA Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-37
ARMA Model with No Constant Term . . . . . . . . . . . . . . . . . 5-38
ARMA Model with Known Parameter Values . . . . . . . . . . . 5-39

ARIMA Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-41

ARIMA Model Specifications . . . . . . . . . . . . . . . . . . . . . . . . . 5-43

Default ARIMA Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-43
ARIMA Model with Known Parameter Values . . . . . . . . . . . 5-44

Multiplicative ARIMA Model . . . . . . . . . . . . . . . . . . . . . . . . . 5-46

Multiplicative ARIMA Model Specifications . . . . . . . . . . . . 5-48

Seasonal ARIMA Model with No Constant Term . . . . . . . . . 5-48
Seasonal ARIMA Model with Known Parameter Values . . . 5-49

Specify Multiplicative ARIMA Model . . . . . . . . . . . . . . . . . . 5-52

ARIMA Model Including Exogenous Covariates . . . . . . . . . 5-58

ARIMAX(p,D,q) Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-58

xiii
 Conventions and Extensions of the ARIMAX Model . . . . . . . 5-58

ARIMAX Model Specifications . . . . . . . . . . . . . . . . . . . . . . . . 5-61

Specify ARIMAX Model Using Name-Value Pairs . . . . . . . . 5-61
Specify ARMAX Model Using Dot Notation . . . . . . . . . . . . . 5-62

Modify Properties of Conditional Mean Model Objects . . . 5-65

Dot Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-65
Nonmodifiable Properties . . . . . . . . . . . . . . . . . . . . . . . . . . 5-69

Specify Conditional Mean Model Innovation Distribution . 5-72

About the Innovation Process . . . . . . . . . . . . . . . . . . . . . . . 5-72
Choices for the Variance Model . . . . . . . . . . . . . . . . . . . . . . 5-73
Choices for the Innovation Distribution . . . . . . . . . . . . . . . . 5-73
Specify the Innovation Distribution . . . . . . . . . . . . . . . . . . . 5-74
Modify the Innovation Distribution . . . . . . . . . . . . . . . . . . . 5-76

Specify Conditional Mean and Variance Models . . . . . . . . . 5-79

Impulse Response Function . . . . . . . . . . . . . . . . . . . . . . . . . . 5-86

Plot the Impulse Response Function . . . . . . . . . . . . . . . . . . 5-88

Moving Average Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-88
Autoregressive Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-89
ARMA Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-91

Box-Jenkins Differencing vs. ARIMA Estimation . . . . . . . . 5-94

Maximum Likelihood Estimation for Conditional Mean

Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-98
Innovation Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-98
Loglikelihood Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-99

Conditional Mean Model Estimation with Equality

Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-101

Presample Data for Conditional Mean Model Estimation . 5-103

Initial Values for Conditional Mean Model Estimation . . . 5-106

Optimization Settings for Conditional Mean Model

Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-108
Optimization Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-108

xiv Contents
 Conditional Mean Model Constraints . . . . . . . . . . . . . . . . 5-111

Estimate Multiplicative ARIMA Model . . . . . . . . . . . . . . . . 5-113

Model Seasonal Lag Effects Using Indicator Variables . . . 5-117

Forecast IGD Rate Using ARIMAX Model . . . . . . . . . . . . . . 5-122

Estimate Conditional Mean and Variance Models . . . . . . . 5-129

Choose ARMA Lags Using BIC . . . . . . . . . . . . . . . . . . . . . . . 5-135

Infer Residuals for Diagnostic Checking . . . . . . . . . . . . . . 5-140

Monte Carlo Simulation of Conditional Mean Models . . . . 5-146

What Is Monte Carlo Simulation? . . . . . . . . . . . . . . . . . . . 5-146
Generate Monte Carlo Sample Paths . . . . . . . . . . . . . . . . . 5-146
Monte Carlo Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-147

Presample Data for Conditional Mean Model Simulation . 5-149

Transient Effects in Conditional Mean Model Simulations 5-150

Simulate Stationary Processes . . . . . . . . . . . . . . . . . . . . . . . 5-151

Simulate an AR Process . . . . . . . . . . . . . . . . . . . . . . . . . . 5-151
Simulate an MA Process . . . . . . . . . . . . . . . . . . . . . . . . . . 5-156

Simulate Trend-Stationary and Difference-Stationary

Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-163

Simulate Multiplicative ARIMA Models . . . . . . . . . . . . . . . 5-169

Simulate Conditional Mean and Variance Models . . . . . . . 5-174

Monte Carlo Forecasting of Conditional Mean Models . . . 5-180

Monte Carlo Forecasts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-180
Advantage of Monte Carlo Forecasting . . . . . . . . . . . . . . . 5-180

MMSE Forecasting of Conditional Mean Models . . . . . . . . 5-181

What are MMSE Forecasts? . . . . . . . . . . . . . . . . . . . . . . . 5-181
How forecast Generates MMSE Forecasts . . . . . . . . . . . . . 5-181
Forecast Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-183

xv
 Convergence of AR Forecasts . . . . . . . . . . . . . . . . . . . . . . . 5-185

Forecast Multiplicative ARIMA Model . . . . . . . . . . . . . . . . 5-191

Forecast Conditional Mean and Variance Model . . . . . . . . 5-196

Conditional Variance Models

6
Conditional Variance Models . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
General Conditional Variance Model Definition . . . . . . . . . . 6-2
GARCH Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
EGARCH Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
GJR Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-6

Specify GARCH Models Using garch . . . . . . . . . . . . . . . . . . . . 6-8

Default GARCH Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-8
Specify Default GARCH Model . . . . . . . . . . . . . . . . . . . . . . 6-10
Using Name-Value Pair Arguments . . . . . . . . . . . . . . . . . . 6-11
Specify GARCH Model with Mean Offset . . . . . . . . . . . . . . 6-15
Specify GARCH Model with Known Parameter Values . . . . 6-15
Specify GARCH Model with t Innovation Distribution . . . . . 6-16
Specify GARCH Model with Nonconsecutive Lags . . . . . . . . 6-17

Specify EGARCH Models Using egarch . . . . . . . . . . . . . . . . 6-19

Default EGARCH Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-19
Specify Default EGARCH Model . . . . . . . . . . . . . . . . . . . . . 6-21
Using Name-Value Pair Arguments . . . . . . . . . . . . . . . . . . 6-22
Specify EGARCH Model with Mean Offset . . . . . . . . . . . . . 6-26
Specify EGARCH Model with Nonconsecutive Lags . . . . . . . 6-27
Specify EGARCH Model with Known Parameter Values . . . 6-28
Specify EGARCH Model with t Innovation Distribution . . . . 6-29

Specify GJR Models Using gjr . . . . . . . . . . . . . . . . . . . . . . . . 6-31

Default GJR Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-31
Specify Default GJR Model . . . . . . . . . . . . . . . . . . . . . . . . . 6-33
Using Name-Value Pair Arguments . . . . . . . . . . . . . . . . . . 6-34
Specify GJR Model with Mean Offset . . . . . . . . . . . . . . . . . 6-38
Specify GJR Model with Nonconsecutive Lags . . . . . . . . . . . 6-39
Specify GJR Model with Known Parameter Values . . . . . . . 6-40

xvi Contents
 Specify GJR Model with t Innovation Distribution . . . . . . . 6-40

Modify Properties of Conditional Variance Models . . . . . . 6-42

Dot Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-42
Nonmodifiable Properties . . . . . . . . . . . . . . . . . . . . . . . . . . 6-45

Specify the Conditional Variance Model Innovation

Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-48

Specify Conditional Variance Model For Exchange Rates . 6-53

Maximum Likelihood Estimation for Conditional Variance

Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-62
Innovation Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-62
Loglikelihood Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-62

Conditional Variance Model Estimation with Equality

Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-65

Presample Data for Conditional Variance Model

Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-67

Initial Values for Conditional Variance Model Estimation . 6-69

Optimization Settings for Conditional Variance Model

Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-71
Optimization Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-71
Conditional Variance Model Constraints . . . . . . . . . . . . . . . 6-74

Infer Conditional Variances and Residuals . . . . . . . . . . . . . 6-76

Likelihood Ratio Test for Conditional Variance Models . . . 6-82

Compare Conditional Variance Models Using Information

Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-86

Monte Carlo Simulation of Conditional Variance Models . . 6-91

What Is Monte Carlo Simulation? . . . . . . . . . . . . . . . . . . . . 6-91
Generate Monte Carlo Sample Paths . . . . . . . . . . . . . . . . . 6-91
Monte Carlo Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-92

Presample Data for Conditional Variance Model

Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-94

xvii
 Simulate GARCH Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-96

Assess EGARCH Forecast Bias Using Simulations . . . . . . 6-103

Simulate Conditional Variance Model . . . . . . . . . . . . . . . . 6-110

Monte Carlo Forecasting of Conditional Variance Models 6-114

Monte Carlo Forecasts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-114
Advantage of Monte Carlo Forecasting . . . . . . . . . . . . . . . 6-114

MMSE Forecasting of Conditional Variance Models . . . . . 6-116

What Are MMSE Forecasts? . . . . . . . . . . . . . . . . . . . . . . . 6-116
EGARCH MMSE Forecasts . . . . . . . . . . . . . . . . . . . . . . . . 6-116
How forecast Generates MMSE Forecasts . . . . . . . . . . . . . 6-117

Forecast GJR Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-122

Forecast a Conditional Variance Model . . . . . . . . . . . . . . . 6-125

Converting from GARCH Functions to Model Objects . . . 6-128

Multivariate Time Series Models

7
Vector Autoregressive (VAR) Models . . . . . . . . . . . . . . . . . . . 7-3
Types of VAR Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
Lag Operator Representation . . . . . . . . . . . . . . . . . . . . . . . . 7-4
Stable and Invertible Models . . . . . . . . . . . . . . . . . . . . . . . . 7-5
Building VAR Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-5

Multivariate Time Series Data Structures . . . . . . . . . . . . . . . 7-8

Multivariate Time Series Data . . . . . . . . . . . . . . . . . . . . . . . 7-8
Response Data Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-8
Example: Response Data Structure . . . . . . . . . . . . . . . . . . . . 7-9
Data Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-11
Partitioning Response Data . . . . . . . . . . . . . . . . . . . . . . . . . 7-11

Multivariate Time Series Model Creation . . . . . . . . . . . . . . 7-14

Models for Multiple Time Series . . . . . . . . . . . . . . . . . . . . . 7-14
Specifying Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-14

xviii Contents
 Specification Structures with Known Parameters . . . . . . . . 7-15
Specification Structures with No Parameter Values . . . . . . 7-16
Specification Structures with Selected Parameter Values . . 7-17
Displaying and Changing a Specification Structure . . . . . . . 7-19
Determining an Appropriate Number of Lags . . . . . . . . . . . 7-19

VAR Model Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-22

Preparing Models for Fitting . . . . . . . . . . . . . . . . . . . . . . . . 7-22
Changing Model Representations . . . . . . . . . . . . . . . . . . . . 7-23
Fitting Models to Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-24
Examining the Stability of a Fitted Model . . . . . . . . . . . . . . 7-25

Convert a VARMA Model to a VAR Model . . . . . . . . . . . . . . 7-27

Convert a VARMA Model to a VMA Model . . . . . . . . . . . . . . 7-29

Fit a VAR Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-33

Fit a VARMA Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-35

VAR Model Forecasting, Simulation, and Analysis . . . . . . . 7-39

VAR Model Forecasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-39
Data Scaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-40
Calculating Impulse Responses . . . . . . . . . . . . . . . . . . . . . . 7-40

Generate Impulse Responses for a VAR model . . . . . . . . . . 7-42

Compare Generalized and Orthogonalized Impulse Response

Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-45

Forecast a VAR Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-50

Forecast a VAR Model Using Monte Carlo Simulation . . . . 7-53

Multivariate Time Series Models with Regression Terms . 7-57

Design Matrix Structure for Including Exogenous Data . . . . 7-58
Estimation of Models that Include Exogenous Data . . . . . . . 7-62

Implement Seemingly Unrelated Regression Analyses . . . . 7-64

Estimate the Capital Asset Pricing Model Using SUR . . . . . 7-74

Simulate Responses of Estimated VARX Model . . . . . . . . . . 7-80

xix
 VAR Model Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-89

Cointegration and Error Correction Analysis . . . . . . . . . . 7-108

Integration and Cointegration . . . . . . . . . . . . . . . . . . . . . . 7-108
Cointegration and Error Correction . . . . . . . . . . . . . . . . . . 7-108
The Role of Deterministic Terms . . . . . . . . . . . . . . . . . . . . 7-110
Cointegration Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-111

Determine Cointegration Rank of VEC Model . . . . . . . . . . 7-114

Identifying Single Cointegrating Relations . . . . . . . . . . . . 7-116

The Engle-Granger Test for Cointegration . . . . . . . . . . . . . 7-116
Limitations of the Engle-Granger Test . . . . . . . . . . . . . . . 7-116

Test for Cointegration Using the Engle-Granger Test . . . . 7-121

Estimate VEC Model Parameters Using egcitest . . . . . . . . 7-126

Simulate and Forecast a VEC Model . . . . . . . . . . . . . . . . . . 7-129

Generate VEC Model Impulse Responses . . . . . . . . . . . . . . 7-138

Identifying Multiple Cointegrating Relations . . . . . . . . . . 7-143

Test for Cointegration Using the Johansen Test . . . . . . . . 7-144

Estimate VEC Model Parameters Using jcitest . . . . . . . . . 7-147

Compare Approaches to Cointegration Analysis . . . . . . . . 7-150

Testing Cointegrating Vectors and Adjustment Speeds . . 7-154

Test Cointegrating Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . 7-155

Test Adjustment Speeds . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-158

xx Contents
 State-Space Models
8
What Are State-Space Models? . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
State-Space Model Creation . . . . . . . . . . . . . . . . . . . . . . . . . 8-6

What Is the Kalman Filter? . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8

Standard Kalman Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8
State Forecasts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9
Filtered States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-10
Smoothed States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-11
Smoothed State Disturbances . . . . . . . . . . . . . . . . . . . . . . . 8-12
Forecasted Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-12
Smoothed Observation Innovations . . . . . . . . . . . . . . . . . . . 8-13
Kalman Gain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-14
Backward Recursion of the Kalman Filter . . . . . . . . . . . . . . 8-14
Diffuse Kalman Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-15

Explicitly Create State-Space Model Containing Known

Parameter Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-17

Create State Space Model with Unknown Parameters . . . . 8-20

Explicitly Create State-Space Model Containing Unknown
Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-20
Implicitly Create Time-Invariant State-Space Model . . . . . . 8-22

Create State-Space Model Containing ARMA State . . . . . . . 8-24

Implicitly Create State-Space Model Containing Regression

Component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-28

Implicitly Create Diffuse State-Space Model Containing

Regression Component . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-30

Implicitly Create Time-Varying State-Space Model . . . . . . 8-32

Implicitly Create Time-Varying Diffuse State-Space Model 8-35

Create State-Space Model with Random State Coefficient . 8-38

Estimate Time-Invariant State-Space Model . . . . . . . . . . . . 8-41

xxi
 Estimate Time-Varying State-Space Model . . . . . . . . . . . . . 8-45

Estimate Time-Varying Diffuse State-Space Model . . . . . . . 8-50

Estimate State-Space Model Containing Regression

Component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-55

Filter States of State-Space Model . . . . . . . . . . . . . . . . . . . . 8-58

Filter Time-Varying State-Space Model . . . . . . . . . . . . . . . . 8-62

Filter Time-Varying Diffuse State-Space Model . . . . . . . . . . 8-68

Filter States of State-Space Model Containing Regression

Component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-76

Smooth States of State-Space Model . . . . . . . . . . . . . . . . . . . 8-80

Smooth Time-Varying State-Space Model . . . . . . . . . . . . . . . 8-84

Smooth Time-Varying Diffuse State-Space Model . . . . . . . . 8-91

Smooth States of State-Space Model Containing Regression

Component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-99

Simulate States and Observations of Time-Invariant State-

Space Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-103

Simulate Time-Varying State-Space Model . . . . . . . . . . . . 8-107

Simulate States of Time-Varying State-Space Model Using

Simulation Smoother . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-112

Estimate Random Parameter of State-Space Model . . . . . 8-116

Forecast State-Space Model Using Monte-Carlo Methods . 8-125

Forecast State-Space Model Observations . . . . . . . . . . . . . 8-133

Forecast Observations of State-Space Model Containing

Regression Component . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-138

xxii Contents
 Forecast Time-Varying State-Space Model . . . . . . . . . . . . . 8-143

Forecast State-Space Model Containing Regime Change in

the Forecast Horizon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-149

Forecast Time-Varying Diffuse State-Space Model . . . . . . 8-156

Compare Simulation Smoother to Smoothed States . . . . . 8-162

Rolling-Window Analysis of Time-Series Models . . . . . . . . 8-168

Rolling-Window Analysis for Parameter Stability . . . . . . . 8-168
Rolling Window Analysis for Predictive Performance . . . . . 8-169

Assess State-Space Model Stability Using Rolling Window

Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-172
Assess Model Stability Using Rolling Window Analysis . . . 8-172
Assess Stability of Implicitly Created State-Space Model . . 8-176

Choose State-Space Model Specification Using Backtesting 8-181

Functions — Alphabetical List

9

Data Sets and Examples

A

Glossary

xxiii
 1

Getting Started

• “Econometrics Toolbox Product Description” on page 1-2

• “Econometric Modeling” on page 1-3
• “Model Objects, Properties, and Methods” on page 1-8
• “Stochastic Process Characteristics” on page 1-20
• “Bibliography” on page 1-27
1 Getting Started

Econometrics Toolbox Product Description

Model and analyze financial and economic systems using statistical methods

Econometrics Toolbox™ provides functions for modeling economic data. You can select
and calibrate economic models for simulation and forecasting. For time series modeling
and analysis, the toolbox includes univariate ARMAX/GARCH composite models with
several GARCH variants, multivariate VARMAX models, and cointegration analysis.
It also provides methods for modeling economic systems using state-space models and
for estimating using the Kalman filter. You can use a variety of diagnostic functions for
model selection, including hypothesis, unit root, and stationarity tests.

Key Features
• Univariate ARMAX/GARCH composite models, including EGARCH, GJR, and other
variants
• Multivariate simulation and forecasting of VAR, VEC, and cointegrated models
• State-space models and the Kalman filter for estimation
• Tests for unit root (Dickey-Fuller, Phillips-Perron) and stationarity (Leybourne-
McCabe, KPSS)
• Statistical tests, including likelihood ratio, LM, Wald, Engle’s ARCH, and Ljung-Box
Q
• Cointegration tests, including Engle-Granger and Johansen
• Diagnostics and utilities, including AIC/BIC model selection and partial-, auto-, and
cross-correlations
• Hodrick-Prescott filter for business-cycle analysis

1-2
 Econometric Modeling

Econometric Modeling
In this section...
“Model Selection” on page 1-3
“Econometrics Toolbox Features” on page 1-3

Model Selection
A probabilistic time series model is necessary for a wide variety of analysis goals,
including regression inference, forecasting, and Monte Carlo simulation. When selecting
a model, aim to find the most parsimonious model that adequately describes your data. A
simple model is easier to estimate, forecast, and interpret.

• Specification tests help you identify one or more model families that could plausibly
describe the data generating process.
• Model comparisons help you compare the fit of competing models, with penalties for
complexity.
• Goodness-of-fit checks help you assess the in-sample adequacy of your model, verify
that all model assumptions hold, and evaluate out-of-sample forecast performance.

Model selection is an iterative process. When goodness-of-fit checks suggest model

assumptions are not satisfied—or the predictive performance of the model is not
satisfactory—consider making model adjustments. Additional specification tests, model
comparisons, and goodness-of-fit checks help guide this process.

Econometrics Toolbox Features

Modeling Features Related Functions
Questions
What is the • The conditional mean and variance models in this • arima
dimension of toolbox are for modeling univariate, discrete data. • egarch
my response • Separate models are available for multivariate,
variable? • egcitest
discrete data, such as VAR and VEC models.
• garch
• State-space models support univariate or
multivariate response variables. • gjr
• jcontest
• ssm

1-3
1 Getting Started

Modeling Features Related Functions

Questions
• vgxpred
• vgxsim
• vgxvarx
Is my time series • Stationarity tests are available. If your data is • arima
stationary? not stationary, consider transforming your data. • i10test
Stationarity is the foundation of many time series
models. • kpsstest
• Or, consider using a nonstationary ARIMA model if • lmctest
there is evidence of a unit root in your data.
Does my time • Unit root tests are available. Evidence in favor • adftest
series have a unit of a unit root suggests your data is difference • arima
root? stationary.
• i10test
• You can difference a series with a unit root until
it is stationary, or model it using a nonstationary • pptest
ARIMA model. • vratiotest
How can I handle • You can deseasonalize (seasonally adjust) your • arima
seasonal effects? data. Use seasonal filters or regression models to • regARIMA
estimate the seasonal component.
• Seasonal ARIMA models use seasonal differencing
to remove seasonal effects. You can also include
seasonal lags to model seasonal autocorrelation
(both additively and multiplicatively).
Is my data • Sample autocorrelation and partial autocorrelation • arima
autocorrelated? functions help identify autocorrelation. • autocorr
• Conduct a Ljung-Box Q-test to test autocorrelations • fgls
at several lags jointly.
• hac
• If autocorrelation is present, consider using a
conditional mean model. • lbqtest
• For regression models with autocorrelated errors, • parcorr
consider using FGLS or HAC estimators. If the • regARIMA
error model structure is an ARIMA model, consider
using a regression model with ARIMA errors.

1-4
 Econometric Modeling

Modeling Features Related Functions

Questions
What if my data • Looking for autocorrelation in the squared • archtest
is heteroscedastic residual series is one way to detect conditional • egarch
(exhibits heteroscedasticity.
volatility • fgls
• Engle’s ARCH test evaluates evidence against
clustering)? the null of independent innovations in favor of an • garch
ARCH model alternative. • gjr
• To model conditional heteroscedasticity, consider • hac
using a conditional variance model.
• For regression models that exhibit heteroscedastic
errors, consider using FGLS or HAC estimators.
Is there an • You can use a Student’s t distribution to model • arima
alternative to fatter tails than a Gaussian distribution (excess • egarch
a Gaussian kurtosis).
innovation • garch
• You can specify a t innovation distribution for all
distribution for conditional mean and variance models, and ARIMA • gjr
leptokurtic data? error models in Econometrics Toolbox. • regARIMA
• You can estimate the degrees of freedom of the t
distribution along with other model parameters.
How do I decide • You can compare nested models using • aicbic
between these misspecification tests, such as the likelihood ratio • lmtest
models? test, Wald’s test, or Lagrange multiplier test.
• lratiotest
• Information criteria, such as AIC or BIC, compare
model fit with a penalty for complexity. • waldtest

Do I have two • The Johansen and Engle-Granger cointegration • egcitest

or more time tests assess evidence of cointegration. • jcitest
series that are • Consider using the VEC model for modeling
cointegrated? • jcontest
multivariate, cointegrated series.
• Also consider cointegration when regressing
time series. If present, it can introduce spurious
regression effects.

1-5
1 Getting Started

Modeling Features Related Functions

Questions
What if I want to • ARIMAX and VARX models are available in this • arima
include predictor toolbox. • ssm
variables? • State-space models support predictor data. • vgxvarx
What if I want • Regression models with ARIMA errors are • fgls
to implement available in this toolbox. • hac
regression, but • Regress robustly using FGLS or HAC estimators.
the classical • mvregress
linear model • For a series of examples on time series regression • regARIMA
assumptions do techniques that illustrate common principles
and tasks in time series regression modeling, see • regress (Statistics
not apply? and Machine
Econometrics Toolbox Examples.
Learning Toolbox)
• For more regression options, see Statistics and
Machine Learning Toolbox™ documentation.
How do use the Standard, linear state-space modeling is available in ssm
Kalman filter to this toolbox.
analyze several
unobservable,
linear, stochastic
time series
and several,
observable,
linear, stochastic
functions of
them?

Related Examples
• “Box-Jenkins Model Selection” on page 3-4
• “Detect Autocorrelation” on page 3-18
• “Detect ARCH Effects” on page 3-28
• “Unit Root Tests” on page 3-44
• “Time Series Regression I: Linear Models”
• “Time Series Regression II: Collinearity and Estimator Variance”
• “Time Series Regression III: Influential Observations”

1-6
 Econometric Modeling

• “Time Series Regression IV: Spurious Regression”

• “Time Series Regression V: Predictor Selection”
• “Time Series Regression VI: Residual Diagnostics”
• “Time Series Regression VII: Forecasting”
• “Time Series Regression VIII: Lagged Variables and Estimator Bias”
• “Time Series Regression IX: Lag Order Selection”
• “Time Series Regression X: Generalized Least Squares and HAC Estimators”

More About
• “Trend-Stationary vs. Difference-Stationary Processes” on page 2-7
• “Box-Jenkins Methodology” on page 3-2
• “Goodness of Fit” on page 3-88
• “Regression Models with Time Series Errors” on page 4-6
• “Nonspherical Models” on page 3-94
• “Conditional Mean Models” on page 5-3
• “Conditional Variance Models” on page 6-2
• “Vector Autoregressive (VAR) Models” on page 7-3
• “Cointegration and Error Correction Analysis” on page 7-108

1-7
1 Getting Started

Model Objects, Properties, and Methods

In this section...
“Model Objects” on page 1-8
“Model Properties” on page 1-9
“Specify Models” on page 1-11
“Retrieve Model Properties” on page 1-16
“Modify Model Properties” on page 1-17
“Methods” on page 1-18

Model Objects

After you have a potential model for your data, you must specify the model to MATLAB®
to proceed with your analysis. Econometrics Toolbox has model objects for storing
specified econometric models. For univariate, discrete time series analysis, there are five
available model objects:

• arima — for integrated, autoregressive, moving average (ARIMA) models optionally

containing exogenous predictor variables.
• garch — for generalized autoregressive conditional heteroscedaticity models
(GARCH)
• egarch — for exponential GARCH models
• gjr — for Glosten-Jagannathan-Runkle models
• regARIMA — for regression models with ARIMA errors

For multivariate, discrete time series analysis you can create a state-space model using
an ssm model obvect.

To create a model object, specify the form of your model to one of the model functions
(e.g., arima or garch). The function creates the model object of the corresponding type in
the MATLAB workspace, as shown in the figure.

1-8
 Model Objects, Properties, and Methods

MATLAB Workspace

Specify Create
Proposed
Model (User)
arima (Software)
arima
variable

You can work with model objects as you would with any other variable in MATLAB. For
example, you can assign the object variable a name, view it in the MATLAB Workspace,
and display its value in the Command Window by typing its name.

This image shows a workspace containing an arima model named Mdl.

Model Properties
A model object holds all the information necessary to estimate, simulate, and forecast
econometric models. This information includes the:

• Parametric form of the model

• Number of model parameters (e.g., the degree of the model)
• Innovation distribution (Gaussian or Student’s t)
• Amount of presample data needed to initialize the model

Such pieces of information are properties of the model, which are stored as fields within
the model object. In this way, a model object resembles a MATLAB data structure
(struct array).

The five model types—arima, garch, egarch, gjr, and regARIMA—have properties
according to the econometric models they support. Each property has a predefined name,
which you cannot change.

1-9
1 Getting Started

For example, arima supports conditional mean models (multiplicative and additive
AR, MA, ARMA, ARIMA, and ARIMAX processes). Every arima model object has these
properties, shown with their corresponding names.

Property Name Property Description

Constant Model constant
AR Nonseasonal AR coefficients
MA Nonseasonal MA coefficients
SAR Seasonal AR coefficients (in a multiplicative model)
SMA Seasonal MA coefficients (in a multiplicative model)
D Degree of nonseasonal differencing
Seasonality Degree of seasonal differencing
Variance Variance of the innovation distribution
Distribution Parametric family of the innovation distribution
P Amount of presample data needed to initialize the AR
component of the model
Q Amount of presample data needed to initialize the MA
component of the model

When a model object exists in the workspace, double-click its name in the Workspace
window to open the Variable Editor. The Variable Editor shows all model properties and
their names.

1-10
 Model Objects, Properties, and Methods

Notice that in addition to a name, each property has a value.

Specify Models
Specify a model by assigning values to model properties. You do not need, nor are you
able, to specify a value for every property. The constructor function assigns default
values to any properties you do not, or cannot, specify.

Tip It is good practice to be aware of the default property values for any model you create.

In addition to having a predefined name, each model property has a predefined data type.
When assigning or modifying a property’s value, the assignment must be consistent with
the property data type.

For example, the arima properties have these data types.

1-11
1 Getting Started

Property Name Property Data Type

Constant Scalar
AR Cell array
MA Cell array
SAR Cell array
SMA Cell array
D Nonnegative integer
Seasonality Nonnegative integer
Variance Positive scalar
Distribution struct array
P Nonnegative integer (you cannot specify)
Q Nonnegative integer (you cannot specify)

Specify an AR(2) Model

To illustrate assigning property values, consider specifying the AR(2) model

where the innovations are independent and identically distributed normal random
variables with mean 0 and variance 0.2. This is a conditional mean model, so use arima.
Assign values to model properties using name-value pair arguments.

This model has two AR coefficients, 0.8 and -0.2. Assign these values to the property AR
as a cell array, {0.8,-0.2}. Assign the value 0.2 to Variance, and 0 to Constant.
You do not need to assign a value to Distribution because the default innovation
distribution is 'Gaussian'. There are no MA terms, seasonal terms, or degrees of
integration, so do not assign values to these properties. You cannot specify values for the
properties P and Q.

In summary, specify the model as follows:

Mdl = arima('AR',{0.8,-0.2},'Variance',0.2,'Constant',0)

Mdl =

1-12
 Model Objects, Properties, and Methods

ARIMA(2,0,0) Model:
--------------------
Distribution: Name = 'Gaussian'
P: 2
D: 0
Q: 0
Constant: 0
AR: {0.8 -0.2} at Lags [1 2]
SAR: {}
MA: {}
SMA: {}
Variance: 0.2

The output displays the value of the created model, Mdl. Notice that the property
Seasonality is not in the output. Seasonality only displays for models with seasonal
integration. The property is still present, however, as seen in the Variable Editor.

1-13
1 Getting Started

Mdl has values for every arima property, even though the specification included only
three. arima assigns default values for the unspecified properties. The values of SAR,
MA, and SMA are empty cell arrays because the model has no seasonal or MA terms.
The values of D and Seasonality are 0 because there is no nonseasonal or seasonal
differencing. arima sets:

• P equal to 2, the number of presample observations needed to initialize an AR(2)

model.

1-14
 Model Objects, Properties, and Methods

• Q equal to 0 because there is no MA component to the model (i.e., no presample

innovations are needed).

Specify a GARCH(1,1) Model

As another illustration, consider specifying the GARCH(1,1) model

where

Assume follows a standard normal distribution.

This model has one GARCH coefficient (corresponding to the lagged variance term) and
one ARCH coefficient (corresponding to the lagged squared innovation term), both with
unknown values. To specify this model, enter:
Mdl = garch('GARCH',NaN,'ARCH',NaN)

Mdl =

GARCH(1,1) Conditional Variance Model:

--------------------------------------
Distribution: Name = 'Gaussian'
P: 1
Q: 1
Constant: NaN
GARCH: {NaN} at Lags [1]
ARCH: {NaN} at Lags [1]

The default value for the constant term is also NaN. Parameters with NaN values need to
be estimated or otherwise specified before you can forecast or simulate the model. There
is also a shorthand syntax to create a default GARCH(1,1) model:
Mdl = garch(1,1)

Mdl =

GARCH(1,1) Conditional Variance Model:

1-15
1 Getting Started

--------------------------------------
Distribution: Name = 'Gaussian'
P: 1
Q: 1
Constant: NaN
GARCH: {NaN} at Lags [1]
ARCH: {NaN} at Lags [1]

The shorthand syntax returns a GARCH model with one GARCH coefficient and one
ARCH coefficient, with default NaN values.

Retrieve Model Properties

The property values in an existing model are retrievable. Working with models resembles
working with struct arrays because you can access model properties using dot notation.
That is, type the model name, then the property name, separated by '.' (a period).

For example, consider the arima model with this AR(2) specification:
Mdl = arima('AR',{0.8,-0.2},'Variance',0.2,'Constant',0);

To display the value of the property AR for the created model, enter:
arCoefficients = Mdl.AR

arCoefficients =

[0.8000] [-0.2000]

AR is a cell array, so you must use cell-array syntax. The coefficient cell arrays are lag-
indexed, so entering
secondARCoefficient = Mdl.AR{2}

secondARCoefficient =

-0.2000

returns the coefficient at lag 2. You can also assign any property value to a new variable:
ar = Mdl.AR

1-16
 Model Objects, Properties, and Methods

ar =

[0.8000] [-0.2000]

Modify Model Properties

You can also modify model properties using dot notation. For example, consider this
AR(2) specification:
Mdl = arima('AR',{0.8,-0.2},'Variance',0.2,'Constant',0)

Mdl =

ARIMA(2,0,0) Model:
--------------------
Distribution: Name = 'Gaussian'
P: 2
D: 0
Q: 0
Constant: 0
AR: {0.8 -0.2} at Lags [1 2]
SAR: {}
MA: {}
SMA: {}
Variance: 0.2

The created model has the default Gaussian innovation distribution. Change the
innovation distribution to a Student's t distribution with eight degrees of freedom. The
data type for Distribution is a struct array.

Mdl.Distribution = struct('Name','t','DoF',8)

Mdl =

ARIMA(2,0,0) Model:
--------------------
Distribution: Name = 't', DoF = 8
P: 2
D: 0
Q: 0

1-17
1 Getting Started

Constant: 0
AR: {0.8 -0.2} at Lags [1 2]
SAR: {}
MA: {}
SMA: {}
Variance: 0.2

The variable Mdl is updated accordingly.

Methods
Methods are functions that accept models as inputs. In Econometrics Toolbox, these
functions accept arima, garch, egarch, gjr, and regARIMA models:

• estimate
• infer
• forecast
• simulate

Methods can distinguish between model objects (e.g., an arima model vs. a garch
model). That is, some methods accept different optional inputs and return different
outputs depending on the type of model that is input.

Find method reference pages for a specific model by entering, for example, doc
arima.estimate.

See Also
regARIMA | ssm | arima | egarch | garch | gjr | struct

Related Examples
• “Specify Conditional Mean Models Using arima” on page 5-6
• “Specify GARCH Models Using garch” on page 6-8
• “Specify EGARCH Models Using egarch” on page 6-19
• “Specify GJR Models Using gjr” on page 6-31

More About
• Using garch Objects

1-18
 Model Objects, Properties, and Methods

• Using egarch Objects

• Using gjr Objects
• “Econometric Modeling” on page 1-3
• “Conditional Mean Models” on page 5-3
• “Conditional Variance Models” on page 6-2

1-19
1 Getting Started

Stochastic Process Characteristics

In this section...
“What Is a Stochastic Process?” on page 1-20
“Stationary Processes” on page 1-21
“Linear Time Series Model” on page 1-22
“Lag Operator Notation” on page 1-22
“Characteristic Equation” on page 1-23
“Unit Root Process” on page 1-24

What Is a Stochastic Process?

A time series yt is a collection of observations on a variable indexed sequentially over
several time points t = 1, 2,...,T. Time series observations y1, y2,...,yT are inherently
dependent. From a statistical modeling perspective, this means it is inappropriate to
treat a time series as a random sample of independent observations.

The goal of statistical modeling is finding a compact representation of the data-

generating process for your data. The statistical building block of econometric time series
modeling is the stochastic process. Heuristically, a stochastic process is a joint probability
distribution for a collection of random variables. By modeling the observed time series yt
as a realization from a stochastic process y = y t ;t = 1,...,T , it is possible to accommodate
{ }
the high-dimensional and dependent nature of the data. The set of observation times T
can be discrete or continuous. Figure 1-1 displays the monthly average CO2 concentration
(ppm) recorded by the Mauna Loa Observatory in Hawaii from 1980 to 2012 [2].

1-20
 Stochastic Process Characteristics

Figure 1-1. Monthly Average CO2

Stationary Processes
Stochastic processes are weakly stationary or covariance stationary (or simply, stationary)
if their first two moments are finite and constant over time. Specifically, if yt is a
stationary stochastic process, then for all t:

• E(yt) = μ < ∞.
•
V(yt) = s 2 < ∞.
• Cov(y , y ) = γ for all lags h π 0.
t t–h h

1-21
1 Getting Started

Does a plot of your stochastic process seem to increase or decrease without bound? The
answer to this question indicates whether the stochastic process is stationary. “Yes”
indicates that the stochastic process might be nonstationary. In Monthly Average CO2,
the concentration of CO2 is increasing without bound which indicates a nonstationary
stochastic process.

Linear Time Series Model

Wold’s theorem [1] states that you can write all weakly stationary stochastic processes in
the general linear form

•
yt = m + Ây ie t- i + e t .
i =1

Here, e t denotes a sequence of uncorrelated (but not necessarily independent) random

variables from a well-defined probability distribution with mean zero. It is often called
the innovation process because it captures all new information in the system at time t.

Lag Operator Notation

The lag operator L operates on a time series yt such that Li yt = yt-i .

An mth-degree lag polynomial of coefficients b1, b2,...,bm is defined as

B( L) = (1 + b1 L + b2 L2 + … + bm Lm ).

In lag operator notation, you can write the general linear model using an infinite-degree
2
polynomial y ( L) = (1 + y 1 L + y 2 L + …),

yt = m + y ( L) et .

You cannot estimate a model that has an infinite-degree polynomial of coefficients with
a finite amount of data. However, if y ( L) is a rational polynomial (or approximately

1-22
 Stochastic Process Characteristics

rational), you can write it (at least approximately) as the quotient of two finite-degree
polynomials.

Define the q-degree polynomial q ( L) = (1 + q1 L + q 2 L2 + … + q q Lq ) and the p-degree

polynomial f ( L) = (1 + f1 L + f2 L2 + … + f p Lp ) . If y ( L) is rational, then

q ( L)
y ( L) = .
f ( L)

Thus, by Wold’s theorem, you can model (or closely approximate) every stationary
stochastic process as

q ( L)
yt = m + et,
f ( L)

which has p + q coefficients (a finite number).

Characteristic Equation
A degree p characteristic polynomial of the linear times series model
yt = f1 yt-1 + f2 yt-2 + ... + f p yt- p + e t is

f ( a) = a p - f1a p-1 - f2 a p-2 - ... - f p .

It is another way to assess that a series is a stationary process. For example, the
characteristic equation of yt = 0.5 yt-1 - 0 .02 yt- 2 + e t is f ( a) = a2 - 0.5 a + 0.02.

The roots of the homogeneous characteristic equation f ( a) = 0 (called the characteristic

roots) determine whether the linear time series is stationary. If every root in f ( a) lies
inside the unit circle, then the process is stationary. Roots lie within the unit circle if

1-23
1 Getting Started

they have an absolute value less than one. This is a unit root process if one or more roots
lie inside the unit circle (i.e., have absolute value of one). Continuing the example, the
characteristic roots of f ( a) = 0 are a = {0.4562,0 .0438}. Since the absolute values of these
roots are less than one, the linear time series model is stationary.

Unit Root Process

A linear time series model is a unit root process if the solution set to its characteristic
equation contains a root that is on the unit circle (i.e., has an absolute value of one).
Subsequently, the expected value, variance, or covariance of the elements of the
stochastic process grows with time, and therefore is nonstationary. If your series has a
unit root, then differencing it might make it stationary.

For example, consider the linear time series model yt = yt-1 + e t , where e t is a white
noise sequence of innovations with variance σ2 (this is called the random walk). The
characteristic equation of this model is z - 1 = 0, which has a root of one. If the initial
t
observation y0 is fixed, then you can write the model as yt = y0 + Â e i . Its expected value
i =1
is y0, which is independent of time. However, the variance of the series is tσ2, which
grows with time making the series unstable. Take the first difference to transform the
series and the model becomes dt = yt - yt-1 = e t . The characteristic equation for this
series is z = 0 , so it does not have a unit root. Note that

• E( dt ) = 0,
which is independent of time,
•
V ( dt ) = s 2 ,
which is independent of time, and
• Cov( dt , dt- s ) = 0,
which is independent of time for all integers 0 < s < t.

Monthly Average CO2 appears nonstationary. What happens if you plot the first
difference dt = yt – yt–1 of this series? Figure 1-2 displays the dt. Ignoring the fluctuations,
the stochastic process does not seem to increase or decrease in general. You can conclude
that dt is stationary, and that yt is unit root nonstationary. For details, see “Differencing”
on page 2-3.

1-24
 Stochastic Process Characteristics

Figure 1-2. Monthly Difference in CO2

References
[1] Wold, H. A Study in the Analysis of Stationary Time Series. Uppsala, Sweden:
Almqvist & Wiksell, 1938.

[2] Tans, P., and R. Keeling. (2012, August). “Trends in Atmospheric Carbon Dioxide.”
NOAA Research. Retrieved October 5, 2012 from http://www.esrl.noaa.gov/gmd/
ccgg/trends/mlo.html.

1-25
1 Getting Started

Related Examples
• “Specify Conditional Mean Models Using arima” on page 5-6
• “Specify GARCH Models Using garch” on page 6-8
• “Specify EGARCH Models Using egarch” on page 6-19
• “Specify GJR Models Using gjr” on page 6-31
• “Simulate Stationary Processes” on page 5-151
• “Assess Stationarity of a Time Series” on page 3-58

More About
• “Econometric Modeling” on page 1-3
• “Conditional Mean Models” on page 5-3
• “Conditional Variance Models” on page 6-2

1-26
Exploring the Variety of Random
Documents with Different Content
back
back
back
Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.

More than just a book-buying platform, we strive to be a bridge

connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.

Join us on a journey of knowledge exploration, passion nurturing, and

personal growth every day!

ebookbell.com