An evaluation forecasting techniques in Kuala Lumpur Stock Exchange (KLSE) finance

The major conflict is regarding the quality of existing literatures in stock market. Evidence shows that some researchers’ supports on incorporating complexity forecasting models while some of them support applied simple forecasting model in forecasting. Up to now the existing studies still far from completed. Hence, it had motivated researchers to find the best and the most accurate volatility forecasting models. This study aims to employ various types of forecasting models into Kuala Lumpur Stock Exchange (KLSE) Finance. This study uses daily volatility of KLSE Finance stock prices from the period 1 January 1991 to 31 December 2010. This aim of this paper is to examine which of the model has the potential and tend to provide the accuracy in forecasting samples. Forecasting models employed in this study include random walk, historical mean model, moving average model and simple regression model. This study uses error statistic to obtain the best forecasting models through the model comparison and rankings. There are four types of error statistic to evaluate the best forecasting models, namely Mean Error (ME); Mean Absolute Error (MAE); Root Mean Square Error (RMSE); and Mean Absolute Percent Error (MAPE). The result of this study shows that simple regression model is the best forecasting model to be implemented into KLSE Finance