Validating the adaptive market hypothesis in the Tunisian stock market

Abstract

We investigated how the adaptive market hypothesis (AMH) offers better explanations for stock return behaviour than the popular efficient market hypothesis (EMH) in the Tunisian Stock Market (TSE). Daily stock index returns from April 1999 to February 2018 were employed. We performed linear and non-linear predictability tests to test whether the TSE undergoes time-varying efficiency. Dummy regression models were performed to determine whether market conditions influence return predictability. Our findings show that the TSE witnessed the era of predictability and unpredictability. We found high return predictability during high volatility period but low predictability during the bull and bear conditions. We submit that the AMH is valid in the TSE.