Sentiment and Stock Returns: A Case for Conventional and Islamic equities in Pakistan

Abstract

The study constructs market sentiment index over the period from August 2009 to June 2019 and examines the causality between market sentiment and returns for conventional and Islamic stocks in Pakistan. Using the firm-level data for all stocks listed on Pakistan Stock Exchange, market sentiment index is constructed as the first principal component of six variables: advances-to-decline, premium on dividends, price-to-earnings, relative strength, money flow and turnover rate. We employ the Vector auto-regression model to examine the two-way causal relationship between investor sentiment and aggregate stock return. Our results show that market sentiment has strong predictive power for subsequent conventional stock returns. Sentiment based trading actions of the investors cause persistence in conventional stock returns for one month; however, as these stocks become overpriced, the price movement reverses in two months’ time. In contrast, we do not find any significant association between market sentiment and Islamic stock returns. Our findings are suggestive of different dynamics and investor behavior in Islamic financial markets of Pakistan and along with the existing literature documenting Islamic stocks performance to be at least as good as the conventional stock can be a comfort to the Muslim Investors and may serve as the catalyst to stimulate the growth of Islamic equities.