Predictability in Stock Returns in an Emerging Market: Evidence from KSE 100 Stock Price Index

Abstract

We investigate the persistence in monthly KSE100 excess stock returns over the Treasury bills rates using non-Gaussian state space or unobservable component model with stable distributions and volatility persistence. Results from our non-Gaussian state space model, which is an improvement over Conard and Kaul (1988), show that the conditional distribution has a stable of 1.748 and normality is rejected even after accounting for GARCH. There exists a statistically significant predictable component in the KSE 100 excess stock returns. The optimal predictor in the unconditional expectation of the series is estimated to be 0.18 percent per annum. An evidence of highly nonconstant scales in different periods of time exhibits a tendency towards stock market crashes which invites remedial policy action.

Volatility Dynamics in an Emerging Economy: Case of Karachi Stock Exchange

“…The mixed response on the superiority of a particular kind of model does not allow the choice of one type of asymmetric model and so the popular EGARCH and GJR-GARCH are estimated. It is important to note here that the work done on modeling volatility in Pakistan has been limited to the use of symmetric models and even though presence of leverage effects was tested for by Kiani (2006) [13], asymmetric models have not been employed on Pakistan to formally account for it.…”

“…They also make attempts to factor into their modeling the non-constant nature of the volatility. While an attempt is made by Kiani (2006) [13] to factor in the presence of leverage effects in the stock return series in the Pakistani case, this has not been followed by application of any formal check to test for the presence of such effects. Hence, the efforts of modeling the volatility in the case of studies on Pakistan is limited to the use of symmetric GARCH models.…”

Weak Form of Efficient Market Hypothesis – Evidence from Pakistan

“…(Gujarati & Porter, 2008 MacKinnon (1996) For daily and weekly returns, the significant value of autocorrelations (Q-statistic) gives another proof regarding the rejection of random walk as the p-value is less than 0.05. The results are significant at very first lag for daily and weekly returns only (Haque et al, 2011;Tahir, 2011;Kiani, 2006 andMobarek, 2000). These functions portray the pattern of sequential reliance in the time series returns.…”

“…His findings evidenced that past data had trends and can be used for prediction of future returns. In same line of market predictability, Kiani (2006) rejected EMH while using non-Gaussian state space or unobservable component model on the KSE-100 Index. Sultan, Madah, and Khalid (2013) compare weak form of efficiency of Karachi and Kuwait stock exchanges over the period from 2005 to 2010.…”

“…Autocorrelation test with assumption that different values at lags are not correlated then they cannot be used for prediction and the series is considered as random or stochastic (Mobarek, 2000;Haque et al, 2011;Tahir, 2011;and Kiani, 2006). While classical liner regression assumes that autocorrelation does not lie in the disturbance u i : i.e.,…”