Modeling and simulation of manufacture sector data in Malaysia with detection of outliers: An ARMA-GARCH approach

Abstract

Manufacturing is one of the major sectors that contribute to the economy of Malaysia. The situation of manufacturing sales, especially rubber gloves received the attention of investors to forecast. However, the pattern of economics exposed to unexpected changes, which called outlier which occurred because of internal or external factors. Consequently, give shock to time series data. The main approach used is the hybrid of Autoregressive Moving Average (ARMA) model and Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model. The effectiveness of the volatility model with the existence of outliers are measured by Monte Carlo method. Besides, to find the best ARMA-GARCH model there are four models produced from different specifications in ARMA(a,b) and GARCH(g, h) models. In this paper, we used 216 monthly price data of Standard Malaysian Rubber Grade 20 (SMR 20) in Malaysia. The validity comparison of diagnostic checking is measured on AIC, AICc, SIC and HQIC. While the forecasting performance evaluated using MSE, RMSE and MAPE. The results of the empirical analysis indicate that the ARMA(2,0)-GARCH(1,2) model is the appropriate model to forecast the price of SMR 20 that were used to the manufacturing of rubber gloves in Malaysia.