Volatility forecasting of real estate stock in Malaysia with smooth transition exponential smoothing

In financial market, volatility forecast has been taking the deliberation of the academics and practitioners over the past decades in different areas of study. Malaysian real estate market has been in the long-run appreciation during years 2000-2013. A reliable volatility forecast in real estate stock market (sector) may provide important information for the central bankers, policymakers, investors, developers and public in decision making process (on real estate). Therefore, this research is to study the volatility forecasting performance of various forecasting models for the Malaysian real estate stocks. Daily returns of 33 Malaysian real estate stocks are used in this study. The forecasting models are ad-hoc methods, generalized autoregressive conditional heteroscedasticity (GARCH) models, and the newly proposed Smooth Transition Exponential Smoothing (STES) methods. Using Mean Absolute Error (MAE) as the evaluation criterion, the newly proposed STES models is found to be the most accurate forecasting model among the comparison models.