North South Business Review, Volume 3, Number 1, July 2009, ISSN 1991-4938 FORECASTING DHAKA STOCK EXCHANGE (DSE) RETURN: AN AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA) APPROACH
ASM Shakil Haider School of Business, North South University, Bangladesh Md. Rezaul Kabir Institute of Business Administration, Dhaka University, Bangladesh ABSTRACT This paper investigates the predictability of Dhaka Stock Exchange (DSE) of Bangladesh by proving the market is not weak form efficient and then predicts the monthly index and the return series by using the Autoregressive Integrated Moving Average (ARIMA) process. Through different formal tests on the dataset, the best fitted model selected was ARIMA (3,1,2) for the index series and ARMA(3,1) for the return series. The forecasted values indicate that the market will remain stable with no extreme shocks in near future extending to 2015 The maximum growth possibility of the market indicated in the model is in the year of 2011 specially from September to November; where 2012 remains moderate; and 2013 to 2015 remains in low growth. For the validity of the forecast the Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error(MAPE) and Theil U statistics are checked. Keywords: Forecasting, Stock index, Stock return, ARIMA, Dhaka Stock Exchange (DSE), Chittagong Stock Exchange (CSE)
In a Weak Form Efficient market there will be no statistical dependence among the price changes or returns of the market which implies that prices are random. If any dependence is found, Weak Form Efficiency hypothesis is violated and one may conclude that it is possible to predict the future market return based on past prices. It has been found that the inefficiency of any emerging market occurs due to the size of the market, thinness of trading and quality of information disclosure (Keane, 1983). Dhaka Stock Exchange (DSE) is a recently organized stock market and the first Stock Exchange of Bangladesh> In this stock exchange low depth of capital and liquidity, high level of volatility and market concentration, thin trading, low turnover exist. Though, some of the earlier studies examining the weak form efficiency of DSE were inconclusive, more recent studies examining weak form inefficiency confirmed market inefficiency. If DSE is weak form inefficient which is also a conclusion of this paper, then prediction of the market is possible. Among the two stock exchanges of Bangladesh, Dhaka Stock Exchange was the first and began trading in 1976 with a paid up capital of Tk 0.138 billion and market capitalization of Tk 0.147 billion which was 0.138% of the GDP (Bepari and Mollik, 2008). In order to control the operation of the stock market and protect the interest of the investors the Security and Exchange Commission (SEC) was formed in 1993. But still the market crashed in1996 and it had a
Forecasting Dhaka Stock Exchange (DSE) Return: An Autoregressive Integrated Moving Average (ARIMA) Approach: Haider and Kabir
prolonged effect through 1999 with cumulative a decline of 83.44% and annual rate of decline of 27.82%. This crash damaged the confidence of the investors. Therefore, a model providing directions for the stock market prediction of stock market index and dependable prediction of the return for DSE may play a vital role in reducing uncertainty in the stock prices, reducing the risk of adverse volatility in the market and helping to form an expectation for the market. According
to Pesaran and Timmermann(1995), stock market forecast is important for both the timing of stock investment and the relative investment desirability among the various sectors in the market. In 1997 the government of Bangladesh undertook Capital Market Development Program (CMDP) supported by ADB. But market suffered shocks again during the year of 2010 and 2011 (in November and December of 2010 the shock was extreme and the index moved up to near 9,000 but just on 2011 it...
Please join StudyMode to read the full document