Is it possible to successfully forecast the real price of crude oil at short horizons?

The aim of the paper is to investigate the possibilities of forecasting the real price of crude oil Brent. The analysis follows a recursive scheme. Forecasting models are developed for monthly data from the period between January 1995 and October 2014, and forecasts are generated for the period between January 2005 and October 2014. A wide range of variables describing real economy, financial processes and energy prices is used in our analysis, which makes it particularly valuable. These variables are used to estimate the ARDL models Forecast accuracy generated by these models is compared with two benchmark models: the naive forecast and the AR(1) model. The results obtained indicate that, at short horizons, certain models generate more accurate forecasts than the benchmark models.