Using Bayesian Approach to Study the Time Varying Correlation among Selected Indices of Tehran Stock Exchange

Abstract

Objective: The effect of the contagion between financial assets is one of the most important challenges that investors are faced with. Financial markets have been severely affected by economic, political, and social events and they are subject to volatility and turmoil that causes financial investors and analysts to be disturbed. Therefore, the main objective of the present study is to investigate the effect of shocks on the return volatility of the selected indices. Methods: The main objective of this study is to use random simulation techniques for GARCH models considering multivariate skewed distributions using the Bayesian approach. For this purpose, we used the daily return data of selected Tehran Stock Exchange (TSE) indices during the period from December 13, 2009 to Jan 29, 2003 and also the flexible types of multivariate distributions that can model both skewedness and fat tail. Results: The results of the Bayesian DCC GARCH (1.1) model showed that the effect of shocks on the return volatility is not the same for the selected indices and this situation implies different risk conditions and uncertainty in the returns of these groups. The results also showed that the stock returns of the automotive group spread more volatility from the previous day to the current day in comparison to Banking group and Petroleum group respectively. Conclusion: Shocks on variables affects the correlations among them. Also, GED distribution was identified as an appropriate distribution in line with the characteristics of the skewedness of the studied indices. Therefore, in estimating the risk of assets and portfolio optimization, one can exploit the main idea used in this study.

Keywords

• contagion
• time varying correlation
• bayesian approach
• conditional variance