Comprehensive Research Among MAX,IVOL And MIN Effect

The core of finance research is asset pricing.For several decades,the asset pricing models have been widely studied,mainly to find the determinants of the expected stock returns.Until now,in the extant literature,hundreds of cross-sectional variables have been documented(e.g.,market beta,size,B/M,profitability,investment,momentum,and reversal,among others),including more recently proposed anomaly variables,such as extreme daily returns(future performance of maximum/minimum daily stock returns in the preceding month(MAX/MIN effect))and their impact on idiosyncratic volatility(IVOL)effect.While much research has been done on developed markets,this thesis contributes to the literature examining these—MAX,MIN,and IVOL—anomalies in the context of emerging markets.More precisely,this thesis comprises four empirical studies that examine MAX,MIN,and IVOL effects for the Pakistani stock market(PSX)between 2003 and 2018.The first study mainly focuses on the predictability of the MAX effect in Pakistan;with in-depth analysis of extreme positive returns with several robustness checks,such as pre-,post-,and ex-crisis subperiod analyses,the existence of a MAX effect across different size groups(i.e.,small,medium,big),the impact of using different promising asset pricing models—Capital Asset Pricing Model(CAPM),Carhart's four-factor model(CH4)and Fama-French's three-,five-,and six-factor models(FF3,FF5,and FF6 respectively)—to compute risk-adjusted returns(alphas).Finally,the role of economic condition—examined using Pakistan's annual growth rate in Gross Domestic Product(GDP)—on the MAX effect's significance is detected.The second study observes the predictability of the extreme negative daily returns(MIN)in Pakistan,where the MIN effect is found confined to the stocks that face limits to arbitrage: a proxy of illiquidity and size.The third study explains the significance of extreme returns,lottery mindset,and idiosyncratic volatility anomalies in the Pakistani stock market.More specifically,this thesis examines whether idiosyncratic volatility(IVOL)is subsumed by the MAX or MIN effects or MAX and MIN effects are subsumed by IVOL in the Pakistani stock market? The fourth study employs the five most promising asset pricing models(i.e.,CAPM,CH4,FF3,FF5,FF6)to estimate abnormal-return(alpha)to test IVOL anomaly in Pakistan.In addition,this thesis attempts to examined a new active anomaly picking strategy that is robust to certain considerations.The abstracts of the four studies are as follows.1.Inspired by the findings of Kumar(2009),Bali,Cakici,and Whitelaw(2011,JFE)examine the MAX effect in the US and find that high-MAX stocks significantly underperform the low-MAX stocks in the following month.This study is the first to examine a similar(MAX)effect in the Pakistani stock market.We find a significant negative maximum daily return(MAX)effect on Pakistan's future stock performance,similar to the U.S.and European evidence.The results remain robust even when the holding period is extended the three or six months.,indicating that it takes a relatively long lag(at least more than six months)in the price adjustment again to fundamental levels in Pakistan.While other proxies for extreme returns(e.g.,skewness and IVOL)play weaker roles,both the MAX and idiosyncratic volatility effects coexist independently: similar to other Asian emerging markets such as China and Korea.It plausibly shows that both MAX and IVOL may have independent information to ascertain Pakistan's stock prices,different from the U.S.and European evidence.The results are robust for both the portfolio-level and firm-level cross-sectional analyses and across subperiods,size groups,and alternative factor-definitions and factor-models.The MAX effect is also robust to controls for book-to-market,size,momentum,short-term reversal,illiquidity,market beta,idiosyncratic skewness,systematic skewness,idiosyncratic volatility,and closing price.Interestingly,contrary to findings reported elsewhere,we find that the MAX effect(gambling behavior in Pakistan)exists only when the overall economy is in an expansion state.A battery of tests suggests that triviality in the MAX effect during economic contraction in Pakistan is driven by the more negative subsequent performance of low-MAX stocks(i.e.,the short-leg):This potential explanation is partially supported by the theoretical model of Palfrey and Wang(2012).2.Kaniel et al.(2008)document that some specific types of investors tend to employ contrarian investing strategy—taking a long position on stocks that face large drops in the prices.This investing behavior could lead to a negative association with subsequent stock returns,referred to as the MIN effect.In this study,the relationship between minimum daily returns(MIN)and future sock performance is examined for the Pakistani stock market during the 2003-2018 period.First,the MIN-return relation is found statistically insignificant,where the magnitude is negative(positive)in the average raw(risk-adjusted)returns,contrary to the findings of Wan(2018)for the Chinese stock market.Second,the average slope of multivariate cross-sectional regression that includes MIN and all other control variables is positive.Third,adding MIN into firm-level cross-sectional regressions reverses the IV-return relation: contrary to Ang et al.(2006,2009),while agrees with Tariq and Ansari(2018).Fourth,the positive MIN effect is statistically significant in subperiods(post and ex-crisis)but insignificant across size groups.Additionally,the extreme negative returns(MIN)effect is restricted to the stocks that face limits to arbitrage,i.e.,proxied by illiquidity and small-size.Furthermore,the MIN effect is only exacerbated when the Pakistani economy is expanding.This evidence partially corroborates the argument of Bali et al.(2011).Overall,the Pakistan stock market can be defined as a market of sparsely diversified investors that prefer lottery-type securities.3.Extant literature forms that the expected stock returns exhibit a positive relation,a negative relation,and no relation with the idiosyncratic volatility(IVOL).Bali et al.(2011)report that the MAX effect(a proxy for lottery-like payoff)reverses the anomalous negative association between idiosyncratic volatility and future stock returns.Similarly,the relationship between idiosyncratic volatility and extreme negative/positive returns is also debatable(inconsistent across different markets).This study is the first to explain both the significance and the inter-relationship of extreme returns and idiosyncratic volatility in the Pakistani stock market.The results show that the robust MAX effect is not subsumed by idiosyncratic volatility(IVOL),minimum daily return(MIN),or other control variables.Based on portfolio strategies,i.e.,long-short IVOL hedge portfolio,the results show a positive return spread between the highest and lowest quantiles sorted on idiosyncratic volatility.However,it disappears in the risk-adjusted returns(alphas are estimated using the most promising asset-pricing models: FF3,CH4,FF5,and FF6).The MIN-return relation in this study is both positive and negative but statistically insignificant in both cases.The control variables include book-to-market(B/M),size,momentum,short-term reversal,illiquidity,market beta,idiosyncratic skewness,systematic skewness,and closing price.The negative MAX-return relationship and positive MIN-return relationship indicate a preference for stocks with lottery-type features(i.e.,lottery mindset)and risk-seeking behavior among Pakistani investors.4.This study further extends the analysis of idiosyncratic volatility by dividing a sample into two categories on the basis of abnormal returns signs in the preceding month: firms with positive abnormal return(i.e.,alpha)and negative abnormal return.The results show that the idiosyncratic risk and future stock returns have a negative relationship in both stocks with negative or positive abnormal return(alpha).This study uses the five most promising asset pricing models to estimate abnormal returns(alpha)and finds approximately similar results that are robust after controlling for other stock return predictors(i.e.,Size,market beta,illiquidity,idiosyncratic skewness,book-to-market,systematic skewness,closing price,momentum,and short-term reversals).Further analysis that involves long-and short-leg of mispricing anomalies of Stambaugh et al.(2015)reveals a negative association between preceding IVOL and subsequent returns on both legs;however,the results are robust and stronger for the short legs of the anomaly portfolios.Lastly,buying(selling)the long(short)leg with the highest IVOL generates statistically and economically significant abnormal returns monthly ranging between 0.28% and 0.55%.More importantly,all of the long-short strategies(HL-HS portfolios)outperform the benchmark portfolios.The results reported in this thesis offer new sights for both researchers and investors,and analysis may be instructive for investors(researchers)having to invest(research)mandates towards other emerging stock markets of similar size and characteristics.