Predicting Trend of Producer Price Index in Malaysian Manufacturing Sector: A Comparative Analysis of Time Series Models

Abstract

This research aims to explore the importance of the Producer Price Index (PPI) in the Malaysian manufacturing sector and develop improved forecasting models for the PPI. The PPI is a crucial economic indicator that measures the average change in prices received by domestic producers for their goods and services. Accurate PPI forecasting is imperative for informed decision-making by policymakers, businesses, and investors. The intricate nature of factors and inherent volatility, however, makes precise prediction challenging. This study employs the Box-Jenkins model and Artificial Neural Network (ANN) as forecasting models for the Malaysian manufacturing sector's PPI data. Evaluation metrics such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE) are utilized to compare forecast performances. The results, as indicated in Table 5, reveal that the ANN model (MAE: 0.4337, MAPE: 0.4080) has the lower value of MAE and MAPE compared to the ARIMA model (MAE: 0.5154, MAPE: 0.4830). Conversely, the ARIMA model (RMSE: 0.6794) exhibits a lower value than the ANN model (RMSE: 1.0212). These findings suggest that the ANN model demonstrates a higher accuracy in forecasting, establishing its superior forecasting performance over the ARIMA model. This research provides critical insights for policymakers, businesses, and investors to adeptly manage risks, formulate effective policies, and make well-informed decisions within the manufacturing sector.