Forecasting oil price volatility using spillover effects from uncertainty indices

• Published in: Finance research letters 2021-10, Vol.42, p.101885, Article 101885
• Main Authors: Chatziantoniou, Ioannis, Degiannakis, Stavros, Delis, Panagiotis, Filis, George
• Format: Article
• Language: English
• Subjects: forecasting accuracy; oil price volatility; spillover effects; Uncertainty
• ISSN: 1544-6123; 1544-6131
• DOI: 10.1016/j.frl.2020.101885

•We consider spillovers between oil price volatility, OVX, and key uncertainty indicators.•We use monthly data for the VIX, US EPU, global EPU, geopolitical risk and partisan conflict indices.•We employ the Diebold and Yilmaz framework to extract net pairwise spillovers between uncertainty indicators and OVX.•All different types of uncertainty are linked to OVX, but only spillovers from US EPU contain significant in-sample predictive information.•Spillovers do not generate real out-of-sample forecasting gains at low sampling frequency. We consider spillovers between oil price volatility and key uncertainty indicators and we extend the applicability of the spillover index beyond economic inference, by generating forecasts of oil price volatility. The paper shows that spillovers do not contain significant predictive information, raising critical questions regarding the usefulness of the spillover index for forecasting exercises at low sampling frequency.