A Component Markov Regime-Switching Autoregressive Conditional Range Model

Abstract

In this paper, we develop a component Markov switching conditional volatility model based on the intraday range and evaluate its performance in forecasting the weekly volatility of the S&P 500 index. We compare the performance of the range-based Markov switching model with that of a number of well established return-based and range-based volatility models, namely EWMA, GARCH and FIGARCH models, the Markov Regime-Switching GARCH model of Klaassen (2002), the hybrid EWMA model of Harris and Yilmaz (2009), and the CARR model of Chou (2005). We show that the range-based Markov switching conditional volatility models produce more accurate out-of-sample forecasts, contain more information about true volatility, and exhibit similar or better performance when used for the estimation of value at risk.