Random Walk Behavior In Pakistani Stock Market
Author: Hafsa Matloob

The aim of this study is to check ―weak form of efficiency‖ of All Pakistani indices in the pandemic. These indices are: KSE-100, KSE-30, KMI-30, and KMI All Share Islamic Index. Daily, weekly, and monthly closing prices for 8 years from 1 June 2013 to 30 June 2021 are used. Different parametric and non-parametric tests are employed with the help of E-view and SPSS i.e., for normality testing: Jarque-Bera (JB) and Kolmogrov-Smironv (KS) test are used, for correlation runs test and autocorrelation (AC) are used, for stationarity testing: Augmented Dickey Fuller (ADF), Phillips-Perron (PP) and Correlogram tests are used. For testing weak form of efficiency and random walk Multi variance ratio (MVR) tests are used in this study. MVR ratio test is used with both assumptions of heteroscesdicity as well as homoscesdicity. JB test observed values for KMIAS, KSE 30, KMI 30 and KSE 100 in daily, ‗weekly‘, and ‗monthly‘ data for period of 8 years are higher than the critical values. P-value for ‗monthly‘ and ‗weekly‘ return series for KMIAS, KSE 30, KMI 30 and KSE 100 are greater than critical value, according to the KS test results show that weekly and monthly data is not normally distributed, but it is normally distributed at a 90% level of confidence. The findings of autocorrelation function and Q-Ljung box statistics shows the existence of autocorrelation in daily returns but in weekly and monthly returns there is no correlation in any lags. So it is concluded that daily returns do not follow random walk for 8 years. Results of Runs test suggests that there is autocorrelation‘ in daily returns for KMIAS, KSE 30 and KMI 30, but no autocorrelation in both daily and weekly returns for KSE 100. Augmented Dicky-Fuller (ADF) and Phillips-Perron (PP) tests are employed for unit root, results of tests showed that the monthly, weekly and daily returns are stationary at level but non-stationary at first difference for all the Pakistani indices. Results of MVR testing reveal that the series of all Pakistani indices do not follow a random walk. Results demonstrate that Pakistani market indices do not follow random walk and it is not weak form efficient, so all investors could profit from the Pakistani market’s expected behavior. Investors can use technical analysis to forecast future prices and plan a good short-term investment strategy. When making decisions about new stock, managers can profit from market timing. Supervisor:- Dr. Ahmad Fraz

Meta Data

Supervisor: Mariam Mohsin

Related Thesis​