Computer Programs and Data to Accompany
State-Space Models with Regime-Switching:
Classical and Gibbs-Sampling Approaches with Applications

 

All programs are written and Copyright ©1998 by Chang-Jin Kim.
Please refer to http://econ.korea.ac.kr/~cjkim/SSMARKOV.htm for conditions of use.
If you have any questions, please e-mail Chang-Jin Kim either at cjkim@korea.ac.kr or changjin@u.washington.edu.


Chapter 3 - State-Space Models and the Kalman Filter

Chapter 4 - Markov-Switching Models

Chapter 5 - State-Space Models with Markov-Switching

Chapter 6 - State-Space Models with Heteroskedastic Disturbances

Chapter 7 - An Introduction to Bayesian Inference and Gibbs Sampling

Chapter 8 - State-Space Models and Gibbs Sampling

Chapter 9 - Markov-Switching Models and Gibbs Sampling

Chapter 10 - State-Space Models with Markov-Switching and Gibbs Sampling

Chapter 11 - Gibbs Sampling and Parameter Uncertainty: Testing for Mean Reversion in Heteroskedastic Data


Return to the top

 

 

 

 

 

 

Chapter 3 - State-Space Models and the Kalman Filter


Application #1: A Decomposition of Real GDP and the Unemployment Rate into Stochastic Trend and Transitory Components

Application #2: An Application of the Time-Varying Parameter Model to Modeling Changing Conditional Variance

Application #3: Stock & Watson's (1991) Dynamic Factor Model of the Coincident Economic Indicators

Programs:

TVP.OPT - A Time-Varying Parameter Model of U.S. Monetary Growth Function: Based on Kim and Nelson (1989)

UC_UNI.OPT - A Univariate Unobserved Components Model of U.S. Real GDP: Based on Clark (1987)

UC_BI.OPT - A Bivariate Unobserved Components Model of U.S. Real GDP and the Unemployment Rate: Based on Clark (1989)

S&W.OPT - A Dynamic Factor Model of Four Coincident Economic Indicators: An Experimental Coincident Index: Based on Stock and Watson (1991)

Data:

TVP.PRN - see program for details

GDP4795.PRN - Real GDP

LHUR.PRN - Unemployment Rate

S&W.PRN - see program for details


Return to chapter list

 

 

 

 

 

 

 

 

Chapter 4 - Markov-Switching Models


Application #1: Hamilton's (1989) Markov-Switching Model of Business Fluctuations

Application #2: A Unit Root in a Three-State Markov-Switching Model of the Real Interest Rate

Application #3: A Three-State Markov-Switching Model of Stock Returns

Programs:

HMT4_KIM.OPT - An AR(4) Model with a Markov-Switching Mean (2-state): Based on Hamilton's (1989) Filter and Kim's (1994) Smoothing

HMT4_DMY.OPT - An AR(4) Model with a Markov-Switching Mean (2-state): Based on Hamilton's (1989) Filter and Kim's (1994) Smoothing (Dummy variables are incorporated for the mean growth rates)

INTR_S3.OPT - A Three-State Markov-Switching Mean-Variance Model of the Real Interst Rate: Based on Garcia and Perron (1996)

STCK_V3.OPT - A Three-State Markov-Switching Variance Model of Stock Returns: Based on Kim, Nelson, and Startz (1997)

HMT_TVP.OPT - An AR(4) Model with a Markov-Switching Mean (2-state) and Time-Varying Transition Probabilities: Based on Filardo (1994)

Data:

GDP4795.PRN - Real GDP

INT_CPUQ.PRN - see program for details

EW_EXCS.PRN - Equal-Weighted Excess Returns

FILARDO.PRN -see program for details


Return to chapter list

 

 

 

 

 

 

 

 

Chapter 5 - State-Space Models with Markov-Switching


Application #1: Sources of Monetary Growth Uncertainty and Economic Activity

Application #2: Friedman's Plucking Model of Business Fluctuations and Implied Business Cycle Asymmetry

Application #3: A Dynamic Factor Model with Markov-Switching: Business Cycle Turning Points and a New Coincident Index

Programs:

KIM_JE0.OPT - not available at this time

KIM_JE1.OPT - A State-Space Representation of Lam's (1990) Gerneralized Hamilton Model and Kim's (1994) Filter(easier version)

TVPMRKF.OPT - Time-Varying-Parameter Model with Markov-Switching Heteroskedasticity: Based on Kim (1993)

TVPM_JNT.OPT - Time-Varying-Parameter Model with Markov-Switching Heteroskedasticity: Based on Kim (1993); A Joint Estimation of the Output Equation and the Monetary Growth Equation

SW_MS.OPT - Dynamic Factor Model with Markov-Switching (A New Coicident Index): An Application of Kim's (1994) Algorithm by Kim and Yoo (1995)

Data:

KIM_JE.PRN - see program for details

TVPMRKF.PRN - see program for details

SW_MS.PRN - see program for details


Return to chapter list

 

 

 

 

 

 

 

 

 

 

Chapter 6 - State-Space Models with Heteroskedastic Disturbances


Application #1: The Link Between the Inflation Rate and Inflation Uncertainty

Application #2: Transient Fads and the Crash of '87 in the U.S. Stock Market

Programs:

TVPGRCH.OPT - A Time-Varying-Parameter Model with GARCH(1,1) Disturbances: Based on Harvey et. al. (1992)

INF_FNL.OPT - A State-Space Model with Markov-Switching Heteroskedasticity: A Model of the Inflation Rate (Large and Infrequent Permanent Shocks to Inflation): Based on Kim (1993)

STCK.OPT - A Fad Model of Stock Returns(Transient Fads): based on Kim and M. Kim (1996)

GARCH.OPT - GARCH(1,1) Model of Stock Returns

STCK_V2.OPT - A Two-State Markov-Switching Model of Stock Returns

Data:

TVPGRCH.PRN - see program for details

GD4795.PRN - GDP Deflator

RSP2692.PRN - see program for details

TURNER3.PRN - see program for details


<Return to chapter list

 

 

 

 

 

 

Chapter 7 - An Introduction to Bayesian Inference and Gibbs Sampling


Programs:

GBS_AR4.PRG - A Gibbs-Sampling Approach to a Linear AR(4) Model

MLE_AR4.PRG - A Maximum Likelihood Estiamtion of a Linear AR(4) Model

GBS_ATO.PRG - A Gibbs-Sampling Approach to a Regression Model with AR(1) Disturbances

MLE_ATO.PRG - A Maximum Likelihood Estiamtion of a Regression Model with AR(1) Disturbances

Data:

GDP4795.PRN - Real GDP


Return to chapter list

 

 

 

 

 

 

 

Chapter 8 - State-Space Models and Gibbs Sampling


Application #1: A Gibbs-Sampling Approach to a Linear Dynamic Factor Model and a New Coincident Index

Program:

SW_GIBS.PRG - A Gibbs-Sampling Approach to Stock and Watson's Dynamic Factor Model of 4 Coincident Economic Indicators: A New Experimental Coincident Index

Data:

FULLDTA.PRN - see program for details


Return to chapter list

 

 

 

 

 

 

 

Chapter 9 - Markov-Switching Models and Gibbs Sampling


Application #1: A Three-State Markov-Switching Variance Model of Stock Returns

Application #2: A Three-State Markov-Switching Mean-Variance Model of the Real Interest Rate

Program:

GIBS_MS0.PRG - A Gibbs-Sampling Approach to an AR(0) Model with a Two-State Markov-Switching Mean, Homoskedastic Disturbances: Multi-Move Gibbs Sampling [For Real GDP Data]

GIBS_S3.PRG - A Gibbs-Sampling Approach to a Three-State Markov-Switching Variance Model of Stock Returns: Multimove Gibbs-Sampling

G_INT_S3.PRG - A Gibbs-Sampling Approach to a Three-State Markov-Switching Mean-Variance AR(2) Model of the Real Interest Rate: Multimove Gibbs-Sampling

Data:

GDP4795.PRN - Real GDP

CRSPD.PRN - see program for details

INT_CPUQ.PRN - see program for details


Return to chapter list

 

 

 

 

 

 

 

Chapter 10 - State-Space Models with Markov-Switching and Gibbs Sampling


Application #1: Business Cycle Turning Points and a New Coincident Index

Application #2: Business Cycle Duration Dependence within a Dynamic Factor Model

Application #3: An Unobserved Component Model of the Long-Run U.S./U.K. Real Exchange Rate with Heteroskedasticity

Programs:

SWMSGIBS.PRG - A Gibbs-Sampling Approach to a Dynamic Factor Model with Markov-Switching: Based on Kim and Nelson (1998)

Data:

FULLDTA.PRN - see program for details


Return to chapter list

 

 

 

 

 

 

 

Chapter 11 -Gibbs Sampling and Parameter Uncertainty: Testing for Mean Reversion in Heteroskedastic Data


VR_HT_OR.PRG - VR Test Based on Gibbs-Sampling-Augmented Randomization, original returns: Based on Kim, Nelson, and Startz (1998)

VR_HT_ST.PRG - VR Test Based on Gibbs-Sampling-Augmented Randomization, standardized returns: Based on Kim, Nelson, and Startz (1998)

Data:

CRSPD.PRN - see program for details


Return to chapter list