A Time Series Analysis of Major Indexes Using GARCH Model with Regime Shifts

Abstract

There has always been a great interest in learning about changes in the volatility patterns of stocks and other time series due to exogenous shocks. Researchers and investors have also been curious to study the effect of unanticipated shocks on persistence of volatility over time. This paper studies three major indexes and utilizes the Iterated Cumulative Sums of Squares (ICSS) algorithm to capture time periods of sudden changes in volatility. The findings suggest that persistence of shocks to volatility is not as high as generally perceived. Volatility persistence declines significantly when regime shifts are combined with a Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model. This paper provides important implications for investors and financial researchers.