In this study, we have attempted to seek evidence for weak-form of market efficiency for KSE 100 Index. Index returns have been studied from 1st January, 1992 to 30th April, 2013. For further analysis, return series has been divided into these groups: 1992-2012, 1992-1994, 1995-1997, 1998-2000, 2001-2003, 2004-2006, 2007-2009, 2010-2012 and 2013. The paper has made use of both Non-Parametric tests (Kolmogrov-Smirnov goodness of fitness test, Runs test and Phillips-Perron test) and Parametric tests (Auto-correlation test, Box-Pierce (Q) statistic test, Ljung and Box (Q) Statistic test, Augmented Dickey-fuller test, Dickey-fuller GLS test, Jarque-Bera test, Kwiatkowski, Phillips, Schmidt and Shin test, Auto-regression and ARIMA model). For further analysis, Runs test has also been run on 20 companies return series for comparison purpose with the results of index return series. In addition, from KSE 30 Index, 20 companies return series based on the free-float of shares have also been analyzed through Runs test to check if increase in numbers of floating shares does increase the randomness in return series or not. To our knowledge, this paper is the first one on KSE 100 Index to study the overall time frame of return series of KSE 100 Index of 22 years with the several random walk and weak-form efficiency tests to ensure the consistency of results; and to compare the results of runs test of index return series with the results of runs test on companies return series from KSE 100 and KSE 30 Indexes. Overall KSE 100 Index has found to be weak-form inefficient, but the last 4 years have shown some signs of efficiency. Companies return series from KSE 30 Index are found to be more random than companies return series from KSE100 Index.
