Islamic calendar anomalies: evidence from Pakistani firm-level data

Most prior research has tested for monthly regularities based on the Gregorian calendar; by contrast, little attention has been given to other calendars based on different religions or cultures. This paper examines Islamic monthly anomalies in a stock market located within a Muslim country - Pakistan. The study employs data for 106 companies listed on the Karachi Stock Exchange (KSE) over the period from 1995 to 2011 and an asymmetric generalized autoregressive conditional heteroscedasticity model to examine whether the mean value and volatility of share returns in the KSE vary with Islamic months. The results from the model offer very little statistical evidence of a monthly seasonal anomaly in average returns, but there is evidence of monthly patterns in the volatility of returns for KSE equities. This finding suggests that investors can formulate an investment strategy and choose a trading time in order to outperform on a risk-adjusted basis.