<![CDATA[Global crude oil prices are a highly volatile economic indicator influenced by various dynamic factors, requiring forecasting methods capable of capturing both linear and non-linear patterns in time series data. This study aims to evaluate the performance of ARIMA, Random Forest, and LSTM models in predicting West Texas Intermediate (WTI) crude oil prices. The dataset consists of 1,316 daily closing price records from January 2020 to December 2024 obtained from Investing.com. The data were divided using a time-based split approach with a ratio of 80% for training and 20% for testing to preserve temporal characteristics. Model evaluation was conducted using MAE, MSE, RMSE, MAPE, and R-squared. The results indicate that LSTM achieved the best performance with an MAE of 2.914, RMSE of 3.66, MAPE of 0.854%, and an R² of 0.8514. In contrast, ARIMA and Random Forest produced negative R² values, indicating limitations in capturing the non-linear dynamics of oil prices. These findings confirm the effectiveness of deep learning approaches for crude oil price forecasting]]>
Copyrights © 2026
