Project Title: Predicting Natural Rubber Prices: An Analysis of ARIMA-GARCH, Exponential Smoothing and LSTM Approaches
The natural rubber industry contributes significantly to Malaysia’s economy, with natural rubber being one of the most important export items. Natural rubber prices are affected by supply, demand, stock and other indirect factors include technological innovation, exchange rate, weather and so on. Predicting the price of natural rubber is critical for many stakeholders, including farmers, refiners, and distributors. It allows them to manage the risk of price fluctuations and hence avoid financial losses. The objective of this study is to examine and compare the effectiveness of several models in forecasting natural rubber prices. The autoregressive integrated moving average – generalized autoregressive conditional heteroscedasticity (ARIMA-GARCH), exponential smoothing (ES), and long short-term memory (LSTM) models were specifically examined. The historical data for natural rubber prices is divided into two sets: training and testing. 90% of this data is used as the training set, while the remaining 10% is used as the testing set. The performance of the predicting models is compared using RMSE, MAE, MAPE and R-squared metrics. The results indicate that the LSTM model outperformed the ARIMA-GARCH and exponential smoothing models across all error metrics. The LSTM model demonstrated the lowest prediction error, making it the most accurate model for forecasting natural rubber prices.