Comparative analysis of the international oil price forecast based on the ARIMA and GARCH models
Shuping Wang
Abstract
Shuping Wang
Abstract
Based on the analysis of impact factors on oil price fluctuations,from January 1986 to December 2010,WTI crude oil price of monthly data,creation of ARIMA and GARCH models for forecasting oil prices.And by extrapolation from January 2011 to April 2012 WTI crude oil price forecast,forecast effect of testing models.Comparative analysis found that in short-term prediction,ARIMA and GARCH model on oil price forecasts are more accurate,but when oil prices due to the major events and large fluctuations,model prediction accuracy of decline in long-term projections,forecasting of GARCH model better than ARIMA models;overall,GARCH models forecasting accuracy than the ARIMA model.Therefore,in the international oil price forecasting,GARCH models are appropriate.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Based on the analysis of impact factors on oil price fluctuations,from January 1986 to December 2010,WTI crude oil price of monthly data,creation of ARIMA and GARCH models for forecasting oil prices.And by extrapolation from January 2011 to April 2012 WTI crude oil price forecast,forecast effect of testing models.Comparative analysis found that in short-term prediction,ARIMA and GARCH model on oil price forecasts are more accurate,but when oil prices due to the major events and large fluctuations,model prediction accuracy of decline in long-term projections,forecasting of GARCH model better than ARIMA models;overall,GARCH models forecasting accuracy than the ARIMA model.Therefore,in the international oil price forecasting,GARCH models are appropriate.
Key concepts: Autoregressive integrated moving average, Autoregressive conditional heteroskedasticity, Econometrics, West Texas Intermediate, Extrapolation, Oil price, Economics, Crude oil