Objective - The crude oil, also known as black gold, is an essential commodity for the sustainability of various industries in the world. Oil prices play an important role in world economy because it causes repercussions. For example, world oil prices plummeted at the end of 2013 and its impact created fluctuations in prices which had affected world economy badly. The aim of this research is to locate a good model that can help to predict oil price fluctuations so that industries can avoid potential negative impacts.

Methodology/Technique - Data of world oil prices from 1987 to 2016 were extracted from West Texas Intermediate (WTI) and Brent Oil sources. A comparative analysis using Empirical Decomposition and Autoregressive Integrated Moving Average (ARIMA) was applied toidentify differences and data were then analysed through SPSS 23. For this research, a set of models based on the smallest MAPE (Mean Absolute Percentage Error) was proposed.

Findings - Results indicate that the Empirical Decomposition was a more appropriate method for predicting oil prices due to the non-linearity of oil price data. In addition, the MAPE also produced a lower error rate than the ARIMA.

Novelty - In this research, world oil price volatility fromWest Texas Intermediate (WTI) and Brent Oil Price data were examined to predict oil price movement for future anticipations.

Type of Paper: Empirical

Keywords: Forecasting, Oil Prices, Autoregressive Integrated Moving Average, ARIMA, Empirical Decomposition, West Texas Intermediate, Brent Oil Price.