Model Development for Predicting the Crude Oil Price: Comparative Evaluation of Ensemble and Machine Learning Methods

Abstract

AbstractThe crude oil market is unstable, and its price is highly volatile. Due to the Covid-19 pandemic, the price of crude oils goes up and down in a short period of time. Future plans and projects’ policies depend directly and indirectly on the future price of crude oil. So, the aim of this study is to predict the price of crude oil by using machine learning and ensemble algorithm, as well as to show the comparison of performance of Ada Boost, Bagging Lasso and Support Vector Regression model. The study uses crude oil price time series data for analysis and to form a model to predict future price. The actual vs. predicted curve is used to show the performance of each algorithm individually. Analysis shows that the ensemble AdaBoost algorithm displays better performance than other algorithms. The result is validated using mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), two accuracy score function variance score, and R2 score. This study will help the stakeholders of the crude oil industry in making decisions and formulating policies based on forecasted crude oil prices.KeywordsCrude oilPrice predictionEnsemble learningMachine learning