Dynamics in Crude Oil Prices: A Markov-Switching Autoregressive Approach Incorporating Sunspot Activity

This paper deals with the dynamic relationship between world crude oil prices and the activities of sunspots in a two-state Markov-Switching Autoregressive (MS-AR) model. The data set used was from February 1994 to May 2023, with crude oil prices and sunspot numbers as an exogenous variable. The MS-AR framework allows for regime-dependent behavior in the intercept and autoregressive coefficients and in the impact of solar activity on oil prices. The result distinguishes two regimes characterized by different degrees of persistence and sensitivity to solar activity. State 1 is highly persistent in the movement of oil prices with a lag 1 coefficient and has a much weaker negative relationship with sunspot activity. On the other hand, State 2 shows lower persistence with a lag 1 coefficient but a stronger negative relationship with sunspots. The transition probabilities denote high stability in both regimes. Asymmetric switching behaviorfavors State1, which is the more persistent state.