Sammanfattning

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

Nyckelord

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

Publicerad i

Statistics
2007, Volym: 41, nummer: 6, sidor: 510-522
Utgivare: Taylor & Francis

SLU författare

• Xie, Yingfu

  • Institutionen för skogsekonomi, Sveriges lantbruksuniversitet
    

UKÄ forskningsämne

Ekonomi och näringsliv
Samhällsvetenskap


Permanent länk till denna sida (URI)

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