Geopolitical risk and oil price volatility: Evidence from Markov-switching model

Abstract

This study explores the predictability of geopolitical risk (GPR) on oil market volatility using autoregressive Markov-regime switching model, and obtains several remarkable findings. First, in-sample results show that high GPR can lead to high fluctuations in oil market. Considering different market states, GPR has different effects. Second, out-of-sample results indicate that GPR has useful information to forecast oil market volatility. Compared to expansions, GPR has a more powerful ability for forecasting oil price volatility during recessions, which are robust to different robustness tests. Third, GPR is effective in long-term forecast horizons. Moreover, geopolitical risk threats and acts are helpful in forecasting oil price volatility, especially geopolitical risk threats.