Markov chain Monte Carlo method and its application to the stochastic volatility model

Abstract

In the time series analysis of asset prices, the stochastic volatility models have recently attracted attentions of many researchers since it clearly describes time-varying variance of asset returns. However, it is difficult to evaluate the likelihood and obtain the maximum likelihood estimators of parameters for such models. We take Bayesian approach and use Markov chain Monte Carlo (MCMC) method to overcome such a problem. We first describe MCMC method and conduct a survey of the literature for its application to the stochastic volatility model. The empirical analysis of stock returns data is also given.