Volatility Modeling of Karachi Stock Exchange

Abstract

Modeling volatility of emerging and developed capital markets of the world has always been an imperative area of research in the field of Statistics, Finance, and Economics with the detection of volatility clustering, time-varying volatility and asymmetric response of volatility to market activities and trends. The present study is attempted to investigate the most important features of Karachi stock exchange (KSE-100 Index) using the data of daily closing index from Jan 1, 1999 to Dec 31, 2008. For empirical investigation of data, ARMA, ARCH, GARCH and EGARCH models are fitted to model the conditional mean and conditional variance of KSE-100 Index. The results indicate that GARCH (1,1) model is the good fittedmodel but not able to capture the “leverage effect”. So, EGARCH is applied to cope with this problem. EGARCH(1,1) model gives the best fit to data as compared to other applied techniques. Detailed analysis of data escort us to report that KSE-100 Index is highly volatile and positive returns are associated with higher volatility than negative returns of equal magnitude and it is clear that past residuals highly influence the current volatility.