Time-frequency dependence and connectedness among global oil markets: Fresh evidence from higher-order moment perspective

Abstract

Investigating the dependence and connectedness among global oil markets is of great significance for cross-market investors and regulators. However, most of the existing studies are confined to lower-order moments and the time domain. This paper is the first to examine the time-frequency dependence and connectedness among global oil markets from the higher-order moment perspective by applying the wavelet coherence method and the newly proposed time-varying parameter vector autoregression-based frequency connectedness approach. The empirical results demonstrate that higher-order moment dependence among oil markets is weaker than return and volatility dependence. In general, Dubai, Minas, and Tapis oil exhibit relatively higher wavelet coherence with Daqing oil at all moments. The lead-lag relationships are heterogeneous during most sample intervals. The total return and volatility connectedness indices are higher than the skewness and kurtosis. The return connectedness mainly occurs in the short term (1–5 days) whereas the volatility, skewness, and kurtosis connectedness occur in the long run (22-Inf days). West Texas Intermediate oil dominates the return, volatility, and skewness connectedness network while Dubai oil dominates the kurtosis connectedness network. Furthermore, the dynamic total, net, and net-pairwise connectedness indices are all time-varying and event-dependent with the higher-order moment connectedness illustrating more volatile features. Several practical implications are provided for various market agents.