ABSTRACT
Asset price volatility plays an important role in financial risk management such as option pricing and Value-at-Risk. Few would dispute the fact that volatility changes over time and there has been a surge in modelling the dynamics of volatility. Many models have been proposed but they can be classified into two. One is the autoregressive conditional heteroskedasticity (ARCH) and generalized ARCH (GARCH) models proposed by [15] and [6] and their extensions. See [7] and [19] for ARCH models. The other is the stochastic volatility (SV) model proposed by [38]. While ARCH-type models can be estimated using the maximum likelihood method, it is difficult to evaluate the likelihood of SV model. There has also been a surge in developing the
in Bayesian
for the SV model instead of the maximum likelihood estimation (see [9]) and [18]). The SV model is also becoming popular in macroeconometrics. For example, see [16], [25] and [33].
