An empirical analysis of the adaptive market hypothesis and investor sentiment in extreme circumstances

The Efficient Market Hypothesis (EMH) has been widely studied in the literature, however there remains no consensus among academics whether markets are efficient or not. Although it was initially thought to hold, the recent explosion of studies that find that markets are not efficient has cast serious doubt on the validity of the EMH. Furthermore, the vast majority of the literature examines the EMH over some predetermined sample period, disregarding the fact that the level of efficiency may change over time and a large sample period may not be efficient or not for the whole period. A new theory that tries to accommodate both these facets is the Adaptive Market Hypothesis (AMH), proposed by Andrew Lo (2004). This theory enables market efficiency and market inefficiencies to co-exist together and market efficiency to evolve over time.The main objective of this thesis is to examine the AMH and stock return behaviour in major stock markets using very long data and determine whether it is a more appropriate model for describing stock market behaviour than the EMH. A five-type classification is proposed to distinguish the differing behaviour of stock returns over time. Daily data is spilt into five-yearly subsamples and investigated in respect of linear and nonlinear time-series tests, three calendar anomalies and the moving average technical rule. The results suggest that the AMH provides a better description of the behaviour of stock returns than the classic EMH. Linked to the AMH is the fact that investors are not rational and investor psychology plays a real role in investor’s decision making. With that in mind, this thesis also examines the level of investor sentiment in stock returns during World War Two in Britain. This is a time period that has not been studied in great detail and provides an opportunity to examine investor sentiment in extreme circumstances. The empirical results show that there was strong negative investor sentiment from major negative events and a strong level of local bias during the period known as the Blitz