Structural changes in the conditional volatility process of stock market returns

The most successful models in modeling time varying volatility are GARCH type models. However, when financial returns exhibit sudden jumps that are due to structural breaks the GARCH models show high volatility persistence, i.e. integrated behavior of the conditional variance. In such situations models in which the parameters are allowed to change over time are more appropriate. This paper compares different GARCH models in terms of their ability to describe structural changes in returns caused by financial crisis at Croatian stock market. Fat-tailed conditional distribution of innovations is assumed. Moreover, state dependent degrees of freedom are assumed to model possible time varying kurtosis. The empirical analysis demonstrates that Markov regime switching GARCH model resolves the problem of excessive persistence and outperforms uni-regime GARCH models in forecasting volatility when sudden switching occurs in response to financial crisis. Regime switching GARCH models allow different speeds of mean reversion of innovation process on different levels of variance in different time periods. The data set used in the study consists of returns of the CROBEX index daily closing prices obtained from Zagreb Stock Exchange.