Nonlinear ARDL approach for asymmetric effects of investor sentiment on asset pricing in an emerging Asian economy: the Malaysian experience

This paper addresses the issues pertaining to asset pricing model in Malaysian stock market, an emerging Asian economy, using monthly data between January 2001 and December 2015 for all the stocks on the main market of Bursa Malaysia and five different investor sentiment proxies (i.e., market-wide indicators). Employing NARDL nonlinear cointegration approach, we examine the causal relationship between stock excess returns and investor sentiment in the integrated Fama-French three-factor model. The empirical results suggest that the investor sentiment is an added risk factor to help explain directly the mispricing component of returns in the Fama-French three-factor model and thus bridging the current research gap between traditional and behavioural asset-pricing theories in Malaysia. Besides, this paper reveals that the stock returns are affected by sentiment in an asymmetric and nonlinear manner in either short- or long-run. In particular, we found that the immensity of positive changes of sentiment is significantly greater than that of negative changes of sentiment towards stock returns. These findings may help finance professionals to perform smart investing strategies using investor sentiment as a contrarian indicator.