Markov Switching Vector Autoregressive Modelling of the Nigerian Stock Price and Oil Price Series

Abstract

This article studied the relationship between stock prices and crude oil prices of Nigeria using a Markov switching model. Certain properties of the stock price series and crude oil price series such as breaks and stationarity, which are necessary before choosing a multivariate time series model for this relationship were investigated. Unit root and cointegration structural break tests were used where evidence of breaks exists. In particular, each of the series was found to be a nonlinear and nonstationary series with evidence of a structural break. The results of the unit root and cointegration tests in the presence of structural breaks indicated evidence of I (1) and no cointegration between the series. Consequently, a Markov switching VAR (MSM(2)-VAR(1)) model with two regimes was fitted to the data having established the suitability of the series to regime switching models. The results showed that high volatility regime occurs when the economy was under recession. Furthermore, there exists a positive relationship between stock prices and crude oil prices during the high volatility regime and a negative relationship during the low volatility regime. Keywords: MS-VAR, VAR, Crude oil prices, Stock prices, Markov Switching, Structural break <w:LsdException Locked="false" Priority="50" Name="Grid Table 5 Dark Accent 1"