The forecasting power of economic policy uncertainty for energy demand and supply

This paper considers using a global Economic Policy Uncertainty (EPU) index to enhance forecasting energy demand and supply. Our demand and supply are measured from the consumption and production of energies (fossil fuels and renewable energies). In this study, the Bayesian Vector Autoregressive (BVAR) model is employed to forecast energy demand and supply more accurately. Overall, the BVAR model that uses the EPU index as a predictor on a long sample makes a better prediction of energy demand. This result indicates a high predictive power of the EPU index in energy demand forecasting. However, it seems that BVAR without EPU provides a poor job at forecasting energy supply. (C) 2021 The Author(s). Published by Elsevier Ltd.