Does oil impact Islamic stock markets ? evidence from MENA countries based on wavelet and markov switching approaches

This paper combines the Wavelet and Markov switching analysis to examine the impact in the volatility of crude oil prices on the Islamic stock market returns of the Middle East and Northern African countries (MENA) over the period of July 2010 to March 2016. Result tend to show that, in all cases, the variables exhibit less coherence in the short run (first sixteen days) except in Jordan, Oman and Qatar. In general, for the entire analyzed period, the colour code shows that the co-movements between series are more persistent in the medium run (32-64 day cycles) and long run (32-64 day cycles). In the short-run, the direction of the contagion cannot be identified. The coherence is only persistent over the medium run (32-64 day cycle) from the period of 2012-2016. In case of co-movement, the Bahrain, Jordan, Saudi Arabia, and the UAE, Islamic stock indices are leading crude oil. This means that if Bahrain, Jordan, Saudi Arabia and the UAE Islamic Stock markets are bullish the oil price rises. However, over the period of 2011-2012, and 2015-2016 crude oil returns were leading the Kuwait Islamic stock index but in the period of 2014-2015, the Kuwaiti Islamic stock index led the crude oil returns. A similar case can be observed during the period of 2015-2016 where the Tunisian Islamic stock index led crude oil return but in 2016, crude oil returns led the Tunisian Islamic Index. Except for Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates over the long-run (128 day cycle), there is no coherence at that period. They move separately which implies that oil price is independent of the bullish or bearish trend of the Islamic Stock Markets of Bahrain, Jordan and Tunisia. Regarding the issue of Markov regime-switching, the results tend to reject the ‘null hypothesis of no regime shifts’ for the stock markets in the ASEAN countries, which means that the time-varying behaviour of these markets is better captured by the nonlinear MS-AR model.