Last modified: 2005 Aug 10
<<To archive home page.

Time Varying Parameters

• Tsay's (1989) univariate TAR specification test and Tsay's (1998) multivariate TAR specification test (Lo 1999)
• Benoit Bellone's library to estimate Multivariate Hidden Markov Models: MSVARlib-v2.0.zip. Be sure to read the copyright. This package includes a PDF working paper, data bases, GAUSS programs. (Local copy.)
• Bruce Hansen's Markov switching code:
  Retrieve from his site or find local copies here.
  jae_92.zip contains two Gauss programs, one for the Markov switching mean AR(p) model of Hamilton (1989), and one for the Markov switching parameter model as presented in Bruce E. Hansen "The likelihood ratio test under non-standard conditions: Testing the Markov switching model of GNP." Journal of Applied Econometrics, (1992 and 1996).
  Hansen Comments: "Some of the above material is based upon work supported by the National Science Foundation under Grants No. SES-9022176, SES-9120576, SBR-9412339, and SBR-9807111. Any opinions, findings, and conclusions, or recommendations expressed in this material are those of the author(s), and do not necessarily reflect the views of the NSF."
• James Hamilton's Markov switching code:
  Retrieve from his site or find local copies here.
  Estimation of Markov switching models by numerical optimization (written for the numerical optimization protocols from GAUSS version 2.0, estimates a pth-order autoregression with K states).
  Programs for estimation of Markov switching models using the EM algorithm (estimates an N-dimensional vector whose mean and covariance matrix change with the state).
  Specification testing of Markov switching models (includes specification tests described in Journal of Econometrics, Jan. 1996 and Journal of Business Jan. 1996).
  Estimation of Markov-switching ARCH models (implements methods used in J. Econometrics, Sept./Oct. 1994).
  Bivariate analysis of SWARCH and Markov-switching autoregression (implements methods used in J. Applied Econometrics, Sept./Oct. 1996).
  Hamilton comments: "The data and software provided above have been developed under research supported by the National Science Foundation under grants SBR-97-07771, SBR-93-08301, SES-89-20752, SES-87-20731. Any opinions, findings and conclusions or recommendations expressed in this material are those of James D. Hamilton and do not necessarily reflect the views of the National Science Foundation."
• State Space and Regime Switching programs written by Chang-Jin Kim of Korea University that cover GARCH estimation as well as estimation of general Markov switching models. (Local copy.)
• Markov-switching models with time-varying transition probabilities using the Gibbs sampler. (Ellison 03). (Local copy.)
• van Norden and Vigfusson's Switching Regime models. (Gauss software and article, see WP 96-3). (Local copy.)
• Ken Train offers some mixed logit routines along with considerable technical resources.
• Endogenous Switching Regression (Wilde Jun98)