Oil prices and E7 stock prices: an asymmetric evidence using multiple threshold nonlinear ARDL model

For examining the short-run and long-run asymmetric effect of oil prices on stock prices, recent literature uses standard nonlinear autoregressive distributed lag model. However, this model decomposes oil price series into partial sum of positive and negative changes only and fails to examine the effect of extreme changes in the oil price series on stock prices. This study, therefore, extends the existing literature by focusing on the emerging seven countries, i.e., Brazil, India, Russia, China, Mexico, Indonesia, and Turkey, and uses multiple threshold nonlinear ARDL model. This extended model helps to examine the asymmetric effect of extremely small to extremely large changes in the oil price series on stock prices. The estimates from standard nonlinear ARDL model indicate that, in the short run, oil prices significantly and asymmetrically affect stock prices in the context of Russia, Indonesia, and India only, whereas in the long run, insignificant effect is found for all sample countries. On the contrary, multiple threshold nonlinear ARDL model supports asymmetric effect in long run and short run for all sample countries where this effect is stronger in short run. Moreover, all diagnostic tests indicate that this extended model enjoys a better fit and is more stable than the traditional models. The findings, based on this model, provide deeper insights on the relationship between oil prices and stock prices and can be used for investors, policymakers, and other stakeholders.