Wavelet Transform and Autoregressive Integrated Moving Average Combination Approach in Crude Oil Prices Forecasting

Authors

  • Nurull Qurraisha Nadiyya Md-Khair Department of Software Engineering, Faculty of Computing Universiti Teknologi Malaysia 81310 Johor Bahru, Johor, Malaysia
  • Ruhaidah Samsudin Department of Software Engineering, Faculty of Computing Universiti Teknologi Malaysia 81310 Johor Bahru, Johor, Malaysia
  • Ani Shabri Department of Science Mathematics, Faculty of Science Universiti Teknologi Malaysia 81310 Johor Bahru, Johor, Malaysia

DOI:

https://doi.org/10.11113/ijic.v8n2.177

Keywords:

Autoregressive Integrated Moving Average, Wavelet Transform, Hybrid Forecasting Method, Crude Oil Spot Prices, Price Forecast

Abstract

This paper proposes a time series forecasting approach combining wavelet transform and autoregressive integrated moving average (ARIMA) to enhance the precision in forecasting crude oil spot prices series. Wavelet transform splits the original prices series into several subseries, then the most appropriate model of ARIMA is established to predict each respective series and finally all series are combined back to get the original series. The datasets for the experiment consist of crude oil spot prices from Brent North Sea (Brent) and West Texas Intermediate (WTI). Single forecasting model ARIMA and several existing forecasting approaches in the literatures are used to measure the performance of the proposed approach by utilizing the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) collected. Final results have depicted that the proposed approach outperforms other approaches with smaller MAE and RMSE values. Ultimately, it is proven that data decomposition, combined with forecasting method can increase the accuracy in time series forecasting.

Downloads

Published

2018-05-31

How to Cite

Md-Khair, N. Q. N., Samsudin, R., & Shabri, A. (2018). Wavelet Transform and Autoregressive Integrated Moving Average Combination Approach in Crude Oil Prices Forecasting. International Journal of Innovative Computing, 8(2). https://doi.org/10.11113/ijic.v8n2.177

Issue

Section

Information Systems