Investigating ICAPM with mean-reverting dynamic conditional correlation: Evidence from an emerging stock exchange

Abstract

Abstract This study examines the rationality of Intertemporal Capital Assets Pricing Model (ICAPM) for Pakistan Stock Exchange (PSX). The examination is a two-phase process: in the initial phase we estimate the dynamic conditional correlations between portfolio returns with market portfolio and the state variables (such as alternative investment opportunity set) by applying mean reverting dynamic conditional correlation (DCC) model proposed by Engle (2002). In the second phase, dynamic covariances are estimated from correlations, which act as explanatory variable in pricing equation proposed by Bali and Engle (2010) estimated by seemingly unrelated regression (SUR). The state variables incorporated in this study are risk free rate (proxied by the rate on Treasury bill) and Gold Prices. All the findings are consistent with the model expectations for the period 2003 to 2016. Risk aversion coefficient (like beta specification of classical CAPM) is positive and significant, risk free rate shows negative while gold prices show positive and significant relationship. Empirical results in this study would be extremely useful for investors of PSX to explore alternative investment opportunity sets which provide hedging strategies to equity portfolio.