Commodity price prediction using neural networks

• Main Author: Seah, Isaac Zhe Hao
• Other Authors: Wang Lipo
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/72925
• Institution: Nanyang Technological University

The use of crude oil plays an essential role in our everyday life; ranging from electricity generation to vehicle petrol. The change in oil prices affect individuals, companies and nations. Therefore, the need for crude oil price prediction arises. In this paper, we visit different supervised learning methods for commodity price prediction. Two different models of Artificial Neural Network(ANN), namely Backpropagation(BP) model and Radial Basis Function(RBF) model, are constructed and evaluated. Furthermore, another form of supervised learning method: Support Vector Network(SVM), is briefly visited. Three different datasets, such as oil spot price and future contract prices, are utilized to analyse the effectiveness of the supervised learning models on various scenarios. To evaluate the data accuracy, statistical modelling and the MATLAB program were applied. This project offers readers a conclusive insight to different ANN and supervised learning models on crude oil price prediction.