No abstract

Conflicting evidence on weak form efficiency of the Dhaka Stock Market appears to stem from the use of monthly versus daily data, structural changes after the 1996 market crash, and the use of tests with or without heteroscedasticity adjustment. Heteroscedasticity-robust tests indicate short-term predictability of share prices prior to the crash, but not afterwards. Although a heteroscedasticity-robust Box-Pierce test was used by Lo and MacKinlay (1989) in their simulations, our study appears to be the first to apply this test to stock prices. Typical rejection of weak-form market efficiency by the usual autocorrelation tests may be reversed by a heteroscedasticity-robust test. Copyright Blackwell Publishers Ltd, 2005.

No abstract

No abstract