Pustaka Ilmiah Universitas Padjadjaran

Abstrak

Forecasting World Oil Price and its Volatility

Fossil fuel like oil is one major commodity that can affect many aspects of economic activities. Forecasting world oil price can be useful in making prediction of its movement in the future. On the other hand, forecasting world oil price volatility plays a fundamental role in terms of oil risk management. There are many techniques that can be utilized in forecasting oil price and its volatility. Volatility models, such as Autoregressive Conditional Heteroscedasticity (ARCH) and Generalized Autoregressive Conditional Heteroscedasticity (GARCH) become two popular methods in analysing the historical behavior of a series. ARCH and GARCH class models can capture the characteristic of a series, like fat tail and leptokurtic, as the existence of volatility clustering. Forecasting world oil price and its volatility can benefit governments and fuel companies. Efficiency in making prediction of oil price and its volatility can be utilized as hedging strategy in terms of oil investment. This research found that technical analysis can be useful in forecasting Brent and West Texas Intermediate (WTI) oil price. We found that Holt-Winters exponential smoothing is better than ARIMA technique in forecasting Brent and WTI oil price. On the other hand, by employing Threshold GARCH, we found that volatility of Brent and WTI oil price are significantly sensitive to bad news in global economy. The presence of unanticipated bad event in previous month produce higher volatility in the next month. Furthermore, by employing Vector Auto Regression technique, specifically from the information of variance decomposition, Brent oil price, on 10 months average, contributes for about 89.9% to WTI oil price movement, while US$/EUR exchange rate contributes 1.36%. WTI oil price itself contributes only 2.94% to Brent oil price movement, while US$/EUR contributes smaller than WTI oil price, that is 0.83% to Brent oil price movement.