Modeling and predicting stock market returns: A case study on dhaka stock exchange of Bangladesh

Abstract

The available information pertaining to stocks should be entirely reflected in an efficient capital market with a view to aiding policy makers
and investors in designing investment strategy. Hence, this paper investigates the time-series behavior of market returns of Dhaka Stock
Exchange (DSE) of Bangladesh. This study also aims to find out the parsimonious model for forecasting monthly market returns of DSE
more accurately. The monthly data of general index were collected from DSE for the period January 2002 to July 2013. Using Relative
Difference method, monthly market returns were calculated. Autoregressive Integrated Moving Average (ARIMA) models were taken into
account to model the behavior of stock market returns. Subsequently, based on Akaike Information Criterion and forecast errors, the
findings of the study vouchsafe that ARIMA (2, 0, 2) can be employed to model and forecast market returns behavior of DSE efficiently.
Finally, the monthly market returns were forecasted using the parsimonious model for the next 24 months and the predicted values fitted the
observed values reasonably well.