ARMA-GARCH Models: Bayes Estimation Versus MLE, and Bayes Non-stationarity Test

Arbeitspapier

We compare small-sample properties of Bayes estimation and maximum likelihood estimation (MLE) of ARMA-GARCH models. Our Monte Carlo experiments indicate that in small sample, the Bayes estimator beats the MLE. We also develop a Bayes method of testing strict stationarity and ergodicity of the conditional variance in the GARCH(1,1) process, near epoch depencenve (NED), and finiteness of unconditional moments of the GARCH(1,1) process by using a Markov chain Monte Carlo (MCMC) mehtod. We apply this method to test these properties in the ARMA-GARCH models of weekly foreign exchange rates.

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: Series: Working Paper ; No. 1996-19

Classification and Topics

• Classification: Wirtschaft ; Bayesian Analysis: General ; Single Equation Models; Single Variables: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models; Diffusion Processes
• Subject: GARCH ; Markov Chain Monte Carlo (MCMC) ; Near Epoch Dependence (NED)

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Nakatsuma, Teruo ; Tsurumi, Hiroki
• Event: Veröffentlichung
• (who): Rutgers University, Department of Economics
• (where): New Brunswick, NJ
• (when): 1996

Further information

• Handle: http://hdl.handle.net/10419/94323
• Last update: 10.03.2025, 11:42 AM CET