Long Memory Behavior in the Returns of Pakistan Stock Market: ARFIMA-FIGARCH Models

Abstract

This study examines the weak-form market efficiency of Pakistan Stock Market namely Karachi Stock Exchange for the period 2010-2013. The efficiency of stock market has tested by using ARFIMA-FIGARCH models estimated under different distribution assumptions as Normal, Student-t, Skewed Student- t and GED distribution. According to findings of study, ARFIMA model do not support long memory behaviour for the stock market returns. However, FIGARCH model indicate that volatility of market returns has long memory. Moreover, in order to test the feature of long memory in the return and volatility of the stock market simultaneously, ARFIMA-FIGARCH models are estimated according to different distributions simultaneously. Predictable structure of volatility of Pakistan Stock Market display that this market is the weak-form market inefficiency. Consequently, it is possible to say that technical analysis related to this stock market may be valid. This implies that it is possible to predict future stock prices and extra ordinary gains could be obtained trading in this market. Keyword: Weak-Form Efficient Market Hypothesis; Long Memory; ARFIMA-FIGARCH model; Volatility. JEL Classifications: C58; C13; G10; G15; G17