Predictive power of oil prices on CDS spread dynamics of oil-producing countries
研究了油价对巴西、俄罗斯等九个产油国主权信用违约互换利差变化的预测能力，发现油价在2010-2024年整体预测力弱，但在2010-2016和2016-2024两个子样本中预测力增强，且系数符号由负转正。
This paper employs predictive regressions to explore the predictability of sovereign Credit Default Swap (CDS) spread dynamics of relevant oil-producing countries. By incorporating oil prices and additional control variables, we predict the rate of CDS spread changes for Brazil, the UK, Malaysia, Norway, Qatar, Russia, Saudi Arabia , the US , and Venezuela. Our findings reveal that (i) the empirical coefficients of determination ( R 2 ) indicate low in-sample predictability for our entire period of analysis (2010–2024), the R 2 increases markedly when dividing the analysis period into more relevant sub-samples (2010–2016 and 2016–2024); (ii) oil prices are not significant predictors for the full period but become significant in many regressions within sub-samples; (iii) for countries where oil prices are significant in both sub-samples, the coefficient sign changes from negative to positive, suggesting that in more recent years, rising (falling) oil prices signal increasing (decreasing) geopolitical risk , positively (negatively) influencing CDS spreads.