Combination of forecasts for the price of crude oil on the spot market

In this paper, we present a combination of three forecast models, ARIMA, exponential smoothing and dynamic regression, in order to predict the West Texas Intermediate (WTI) crude oil spot price and the Brent North Sea (Brent) crude oil spot price. Using samples from the period between January 1994 and June 2012 (in-sample), we identify the parameters and estimate the models. The validated models are combined to perform a forecast out-of-sample between July 2012 and June 2013. The results demonstrate that among the three models tested in-sample with Brent Prices, based on the MAPE measurement error, the ARIMA (2,1,8) model produced the best result, and the dynamic regression model was the best in-sample model for the WTI price. In the validation phase, the dynamic regression models did not prove to be valid, and therefore the combinations are performed only with the ARIMA and exponential smoothing models. For both proxies of oil prices, the combination of forecasts using ARIMA and exponential smoothing (out-of-sample) performed better than individual ARIMA and exponential models and also better than our benchmark models (naive forecast and Neural Network model). Based on the results, it can be inferred that using to a combination of forecasts to predict WTI and Brent spot prices is promising. We also point out that the selected model is easily replicable in spreadsheets and forecasting software and is based only on the past or lagged WTI or Brent values for future predictions.

[1]  Paul Stevens,et al.  The determination of oil prices 1945–1995 , 1995 .

[2]  Kurt Hornik,et al.  Time Series Analysis and Computational Finance , 2015 .

[3]  John M. Rose,et al.  Forecasting automobile petrol demand in Australia: An evaluation of empirical models , 2010 .

[4]  S. Shapiro,et al.  An Analysis of Variance Test for Normality (Complete Samples) , 1965 .

[5]  G. Cifarelli,et al.  Oil Price Dynamics and Speculation: A Multivariate Financial Approach , 2008 .

[6]  John L. Kling,et al.  A comparison of multivariate forecasting procedures for economic time series , 1985 .

[7]  S. Moshiri,et al.  Forecasting Nonlinear Crude Oil Futures Prices , 2006 .

[8]  Ronald Hagen,et al.  How is the international price of a particular crude determined , 1994 .

[9]  R. Clemen.  Combining forecasts: A review and annotated bibliography , 1989 .

[10]  Perry Sadorsky,et al.  Modeling and forecasting petroleum futures volatility , 2006 .

[11]  Michael Ye,et al.  Forecasting crude oil spot price using OECD petroleum inventory levels , 2002 .

[12]  Siddhivinayak Kulkarni,et al.  Forecasting Model for Crude Oil Price Using Artificial Neural Networks and Commodity Futures Prices , 2009, ArXiv.

[13]  T. Andersen.  THE ECONOMETRICS OF FINANCIAL MARKETS , 1998, Econometric Theory.

[14]  David S. Stoller,et al.  Contrôle estatístico de qualidade , 1966 .

[15]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[16]  Jeffrey M. Wooldridge,et al.  Introductory Econometrics: A Modern Approach , 1999 .

[17]  K. Lai,et al.  Forecasting crude oil price with an EMD-based neural network ensemble learning paradigm , 2008 .

[18]  Fotios Petropoulos,et al.  forecast: Forecasting functions for time series and linear models , 2018 .

[19]  César Augusto Sánchez Correa,et al.  Manual de Econometría , 1994 .

[20]  Peter R. Winters,et al.  Forecasting Sales by Exponentially Weighted Moving Averages , 1960 .

[21]  Jose Alvarez-Ramirez,et al.  Symmetry/anti-symmetry phase transitions in crude oil markets , 2003 .

[22]  José Luis Duarte Ribeiro,et al.  Previsao de demanda: Uma aplicacao dos modelos Box-Jenkins na area de assistencia tecnica de computadores pessoais , 2003 .

[23]  Michael Ye,et al.  A monthly crude oil spot price forecasting model using relative inventories , 2005 .

[24]  Chung-Ming Kuan,et al.  Forecasting exchange rates using feedforward and recurrent neural networks , 1992 .

[25]  Thomas R. Willemain.  The effect of graphical adjustment on forecast accuracy , 1991 .

[26]  Rob J Hyndman,et al.  Another look at measures of forecast accuracy , 2006 .