Nonlinear common trends for the global crude oil market: Markov-switching score-driven models of the multivariate t-distribution

Relevant works from the literature on crude oil market use structural vector autoregressive(SVAR) models with several lags to approximate the true model for the variables change in globalcrude oil production, global real economic activity and log real crude oil prices. Those variables involveseasonality, co-integration, structural changes, and outliers. We introduce nonlinear Markov-switchingscore-driven models with common trends of the multivariate t-distribution (MS-Seasonal-t-QVAR), forwhich filters are optimal according to the Kullback-Leibler divergence. We find that MS-Seasonal-t-QVAR provides a better approximation of the true data generating process and more precise short-runand long-run impulse responses than SVAR