Behavior of calendar anomalies, market conditions and adaptive market hypothesis : evidence from Pakistan stock exchange

The current study investigates Adaptive Market Hypothesis (AMH) via five different calendar effects in Pakistan stock market. For the purpose we examine daily returns of KSE-100 index. The sample comprises 24 years over the period from January 1992 to December 2015. We use sub-sample analysis by utilizing eight different subsamples. Each subsample comprises of equal length of observations (three years each). We further examine subsamples to determine which market condition proves more conduce to the performance of these anomalies. Through the study we enhance the existing literature on AMH as the study first time links both Gregorian and Islamic calendar anomalies with Adaptive Market Hypothesis that permits the performance of well-known calendar effects to fluctuate through time. Furthermore, the study first time links the varying behavior of calendar effects with different conditions prevail in the market to determine what market conditions prove to be most favorable for the performance of these anomalies. We find that behavior of all five calendar anomalies evolve over time as their performance vary from time to time and support the AMH. Furthermore, we also find the presence of the five calendar anomalies under study in different proposed market conditions which indicate varying degree of behavior of calendar anomalies in various market conditions. Overall findings of the current study propose that AMH well clarifies the behavior of calendar effects than traditional Efficient Market Hypothesis (EMH).