Authors: Jarka Arefin, Rashedur M. Rahman
Addresses: Department of Electrical Engineering and Computer Science, North South University, Dhaka, Bangladesh. ' Department of Electrical Engineering and Computer Science, North South University, Dhaka, Bangladesh
Abstract: The Efficient-Market Hypothesis (EMH) asserts that efficient markets are informationally efficient or all information (market, public or private) should reflect on stock prices. No one could earn excess profit using any kind of information in efficient market. There are three forms of efficiency in markets: strong, semi-strong and weak. We tested EMH for Dhaka Stock Exchange (DSE) for the period 2003–2005. We used the excess return market model to test the semi-strong form efficiency of DSE. Two forecasting techniques, Autoregressive Integrated Moving Average (ARIMA) and neural network, are used to test the weak form efficiency of DSE. We get excess return for many stocks listed in DSE, demonstrating that DSE is not an efficient market in semi-strong form. Besides, the DSE market index is not random and the trend could be captured by ARIMA and neural network techniques. Therefore, the DSE is also not an efficient market in weak form.
Keywords: EMH; efficient-market hypothesis; forecasting; econometric methods; neural networks; CAPM; capital asset pricing models; Dhaka; stock exchanges; Bangladesh; informationally efficiency; market information; public information; private information; stock prices; shares; excess profits; strong efficiency; semi-strong efficiency; weak efficiency; excess returns; autoregressive integrated moving average; ARIMA; market indexes; financial services management.
International Journal of Financial Services Management, 2011 Vol.5 No.1, pp.1 - 20
Available online: 28 Jan 2011 *Full-text access for editors Access for subscribers Purchase this article Comment on this article