Crude Palm Oil Price Prediction Using Machine Learning and Hybrid Ensemble Approach

Crude oil is a crucial global commodity. With an increasing population and expanding industries, crude oil prices are highly erratic and subject to high uncertainty. The slightest change in supply or demand can cause significant price changes. This research paper investigates the development of an ensemble model for predicting the price of crude oil. The study builds upon previous research in the field of crude oil price prediction, which emphasizes the limitations of traditional econometric models and the advantages of deep learning algorithms such as Long-Short-Term Memory (LSTM), Convolutional Neural Networks (CNN), and Wavelet Neural Network (WNN). The proposed ensemble model incorporates multiple machine learning algorithms, including Support Vector Regression (SVR), Extreme Gradient Boost (XGB), and Light Gradient Boost (LGB) to enhance prediction speed while maintaining or improving the accuracy. The result shows improvement regarding the prediction speed while giving less error than the previous model.