Publicação
Testing for reverse grander causality between geopolitical risk and oil prices
| Resumo: | This dissertation investigates whether geopolitical risk Granger-causes crude oil returns and volatility, or whether oil market dynamics Granger-causes geopolitical shocks. Using daily and monthly data from 1990 to 2025, this work combines a news-based index of geopolitical risk (GPR) and a market-based global common volatility measure (CoVol), alongside Brent and WTI spot prices, S&P 500 returns and the VIX as control variables.Granger causality tests are performed at both frequencies, followed by the estimation of a parsimonious VAR model with a single lag, and is complemented by robustness checks across oil benchmarks, uncertainty measures and data frequencies. At the daily level, Granger causality tests and VAR estimations reveal contrasting dynamics across uncertainty measures. Daily changes in geopolitical risk (GPR) do not significantly affect Brent or WTI returns, nor do oil prices feed back into geopolitical risk, reflecting the noisy and transitory nature of high-frequency news-based indicators. By contrast, oil prices and common volatility measured by CoVol SQ-1 exhibit stronger short-run interactions, with evidence of bidirectional feedback.At the monthly horizon, geopolitical risk becomes a statistically and economically significant predictor of both Brent and WTI returns, whereas CoVol SQ-1 shows weak and benchmark-specific effects. These results highlight that geopolitical risk and market-wide volatility capture distinct dimensions of uncertainty operating at different time horizons. |
|---|---|
| Autores principais: | Ferreira, João Pedro Lopes Teixeira |
| Assunto: | Geopolitical risk Common global volatility Oil market dynamics Vector autoregression Brent and WTI Granger causality |
| Ano: | 2026 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Católica Portuguesa |
| Idioma: | inglês |
| Origem: | Veritati - Repositório Institucional da Universidade Católica Portuguesa |
| Resumo: | This dissertation investigates whether geopolitical risk Granger-causes crude oil returns and volatility, or whether oil market dynamics Granger-causes geopolitical shocks. Using daily and monthly data from 1990 to 2025, this work combines a news-based index of geopolitical risk (GPR) and a market-based global common volatility measure (CoVol), alongside Brent and WTI spot prices, S&P 500 returns and the VIX as control variables.Granger causality tests are performed at both frequencies, followed by the estimation of a parsimonious VAR model with a single lag, and is complemented by robustness checks across oil benchmarks, uncertainty measures and data frequencies. At the daily level, Granger causality tests and VAR estimations reveal contrasting dynamics across uncertainty measures. Daily changes in geopolitical risk (GPR) do not significantly affect Brent or WTI returns, nor do oil prices feed back into geopolitical risk, reflecting the noisy and transitory nature of high-frequency news-based indicators. By contrast, oil prices and common volatility measured by CoVol SQ-1 exhibit stronger short-run interactions, with evidence of bidirectional feedback.At the monthly horizon, geopolitical risk becomes a statistically and economically significant predictor of both Brent and WTI returns, whereas CoVol SQ-1 shows weak and benchmark-specific effects. These results highlight that geopolitical risk and market-wide volatility capture distinct dimensions of uncertainty operating at different time horizons. |
|---|