Template-Type: ReDIF-Paper 1.0 Series: Tinbergen Institute Discussion Papers Creation-Date: 2010-04-27 Number: 10-046/4 Author-Name: David Ardia Author-Workplace-Name: aeris CAPITAL AG, and University of Fribourg, Switzerland Author-Name: Lennart F. Hoogerheide Author-Workplace-Name: Erasmus University Rotterdam Title: Efficient Bayesian Estimation and Combination of GARCH-Type Models Abstract: This discussion paper resulted in a chapter in: (K. Bocker (Ed.)) 'Rethinking Risk Measurement and Reporting - Volume II: Examples and Applications from Finance', 2010, London: Riskbooks.

This paper proposes an up-to-date review of estimation strategies available for the Bayesian inference of GARCH-type models. The emphasis is put on a novel efficient procedure named AdMitIS. The methodology automatically constructs a mixture of Student-t distributions as an approximation to the posterior density of the model parameters. This density is then used in importance sampling for model estimation, model selection and model combination. The procedure is fully automatic which avoids difficult and time consuming tuning of MCMC strategies. The AdMitIS methodology is illustrated with an empirical application to S&P index log-returns. Several non-nested GARCH-type models are estimated and combined to predict the distribution of next-day ahead log-returns. Classification-JEL: C11, C15, C22 Keywords: GARCH, marginal likelihood, Bayesian model averaging, adaptive mixture of Student-t distributions, importance sampling File-Url: https://papers.tinbergen.nl/10046.pdf File-Format: application/pdf File-Size: 307735 bytes Handle: RePEc:tin:wpaper:20100046