Article,

Analysis of time series subject to changes in regime

.
Journal of Econometrics, 45 (1-2): 39--70 (August 1990)
DOI: 10.1016/0304-4076(90)90093-9

Abstract

This paper introduces an EM algorithm for obtaining maximum likelihood estimates of parameters for processes subject to discrete shifts in autoregressive parameters, with the shifts themselves modeled as the outcome of a discrete-valued Markov process. The simplicity of the EM algorithm permits potential application of the approach to large vector systems.

Tags

Users

  • @becker
  • @alexv
  • @smicha

Comments and Reviews