Forecasting Malaysian Rubber Prices Using Time Series Models
Zeng Zhu
School of Quantitative Sciences, Universiti Utara Malaysia, 06010 UUM, Sintok, Kedah, Malaysia.
Muhammad Bin Mat Yusof *
School of Quantitative Sciences, Universiti Utara Malaysia, 06010 UUM, Sintok, Kedah, Malaysia.
*Author to whom correspondence should be addressed.
Abstract
This study is based on Malaysian rubber price data from January 2016 to December 2024, aiming to forecast the future trend of rubber prices in Malaysia for the period 2025 to 2026 using time series models such as ARIMA and Exponential Smoothing. Forecasting accuracy was systematically compared across models using error evaluation metrics including Mean Absolute Error (MAE), Mean Square Error (MSE), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). The results indicate that the ARIMA (1,1,3) model performs best across all evaluation metrics, allowing it to more precisely capture both the long-term trend and short-term fluctuations of rubber prices. The forecasting results suggest that rubber prices will fluctuate between RM800 and RM900 during 2025 to 2026. This study not only provides a scientific modeling approach and reliable empirical results for price forecasting in the Malaysian rubber industry but also offers valuable data support and decision-making references for policymakers, producers, researchers, and other stakeholders in formulating response strategies, optimizing resource allocation, and mitigating market risks.
Keywords: Malaysia rubber price, time series analysis, forecasting model, ARIMA, exponential smoothing method, double moving average