How do correlations of crude oil prices co-move? A grey correlation-based wavelet perspective

Previous research on the oil market has focused mainly on the static relationship between bivariate oil prices, ignoring the dynamic correlation of bivariate or multivariate oil prices. This study provides a novel perspective on multivariate dynamic correlations for studying the oil market by using an optimal wavelet analysis on the basis of grey correlation. We used China-Daqing and its three reference benchmark oil prices (Brent, Dubai and Minas) as empirical data. Our main findings are as follows. First, the time–frequency phenomena of the analysis results from one-to-one and many-to-one correlation time series support the hypothesis of the regional and global characteristics of the oil market, respectively. Second, the U-shaped wavelet variance plot indicates that the fluctuation intensity of the shortest and longest time–frequency domains plays a leading role in the dynamic process of oil price correlation. For the Chinese government, the oil price adjustment strategy in the short term should reduce the reference weights of Brent, and the long-term strategy should reduce the reference weights of Minas to avoid the risk of a single reference. The investor's portfolio management should pay more attention to the leading oil price of the corresponding period to make clear market timing. Third, the significant lead–lag relationships of oil price correlations showed a time-varying spread phenomenon of benchmark oil prices' relative influence on Daqing, which provides a useful time reference when crafting an oil price adjustment strategy and intertemporal arbitrage.