Can adaptive market hypothesis explain the existence of seasonal anomalies? : evidence from Dhaka Stock Exchange, Bangladesh

This paper examines the behavior of seasonal anomalies in Dhaka Stock Exchange (DSE) of Bangladesh and whether the time varying nature of the anomalies is in line with Adaptive Market Hypothesis (AMH). With this aim the research investigated whether the changes in market conditions, for example: up and down market states, stock market bubbles and crashes, initiation of automated trading system and circuit breaker system can affect the behavior of calendar anomalies and therefore, can provide justification for the seasonal patterns in DSE. To achieve the stated objectives, this study utilizes daily general index values of DSE from 1993 to 2018, with GARCH (1,1) model, Markov switching model, subsample analysis and rolling window analysis. The findings support the existence of AMH at DSE in the form of time-varying nature of seasonal anomalies. However, not all seasonal anomalies examined in the study were found to grow weaker over time. The most important finding of this study is that the investors in emerging stock markets, for example DSE, may not learn from the past investment experiences and show the adapting ability towards changed market conditions in the same manner like the investors in a developed market.