Abstract
In this section we discuss the use of simulation techniques to estimate and forecast MS-VAR processes. A general feature of MS-VAR models is that they approximate non-linear processes as piecewise linear by restricting the processes to be linear in each regime. Since the distribution of the observed variable y t is assumed normal conditional on the unobserved regime vector ξt, the MS-VAR model is well suited for Gibbs sampling techniques.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
For mixtures of normal distributions, Hamilton [1991a] proposed a quasi-Bayesian estimation. However, this is not implemented as a Monte Carlo Chain method, but as a modification of the EM algorithm which has been discussed in Section 6.5.3.
For their MS(2)-ARX(p) model, McCulloch & Tsay [1994b] propose to use the estimates from a linear multiple regression (M = 0) as initial parameter values.
For univariate time series, see Albert & Chib [1993, p.5].
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Krolzig, HM. (1997). Multi-Move Gibbs Sampling. In: Markov-Switching Vector Autoregressions. Lecture Notes in Economics and Mathematical Systems, vol 454. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-51684-9_9
Download citation
DOI: https://doi.org/10.1007/978-3-642-51684-9_9
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63073-9
Online ISBN: 978-3-642-51684-9
eBook Packages: Springer Book Archive