Abstract

Regime-switching GARCH (generalized autoregressive conditionally heteroscedastic) model incorporates the idea of Markov switching into the somehow restrictive GARCH model, which signi¯cantly extends GARCH models. However, the statistical inference for this model is rather di±cult due to the dependence to the whole regime path. In this paper, we obtain the consistency of the quasi-maximum likelihood estimators, by transforming it to an infinite order ARCH model. Simulation studies to illustrate asymptotic behavior of the estimators and a model specification problem are presented

Keywords

Regime-switching; GARCH model; Quasi-MLE; Consistency; Asymptotic normality

Published in

Statistics
2007, Volume: 41, number: 6, pages: 510-522
Publisher: Taylor & Francis

SLU Authors

• Xie, Yingfu

  • Department of Forest Economics, Swedish University of Agricultural Sciences
    

UKÄ Subject classification

Economics and Business
Social Sciences


Permanent link to this page (URI)

https://res.slu.se/id/publ/15205