Title: Forecasting the real price of oil using online search data

Authors: Dean Fantazzini; Nikita Fomichev

Addresses: Moscow School of Economics, 119992, Leninskie Gory, 1, Building 61, M.V. Lomonosov MSU, Moscow, Russia ' Faculty of Economics, Higher School of Economics, Moscow, Russia

Abstract: New models to forecast the real price of oil on the basis of macroeconomic indicators and Google search data are proposed. A large-scale out-of-sample forecasting analysis comparing the different models is performed. It is found that models including both Google data and macroeconomic aggregates statistically outperform the competing models in the short term, while multivariate models including only Google data perform best also for medium and long term forecasts up to 24 months ahead. This finding is confirmed by different robustness checks.

Keywords: oil prices; real price of oil; oil price forecasting; crude oil inventories; global real activity; refiner acquisition cost; multivariate modelling; macroeconomic indicators; Google search data.

DOI: 10.1504/IJCEE.2014.060284

International Journal of Computational Economics and Econometrics, 2014 Vol.4 No.1/2, pp.4 - 31

Received: 16 May 2013
Accepted: 26 Sep 2013

Published online: 08 Apr 2014 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article