دانلود رایگان مقاله انگلیسی شوک های نفتی و نوسانات بازده سهام - الزویر 2018

abstract

Asset return volatility is important to the macroeconomy. This paper asks whether oil price volatility can be used as a predictor of stock return volatility. In contrast with previous research, we focus on the out-of-sample predictive power of oil price volatility rather than on in-sample inference. Formal tests of out-of-sample predictive ability find no evidence supporting the use of oil price volatility as a predictor of future stock return volatility. Further analysis using rolling window estimation and structural break tests shows that the coefficients of this relationship are very unstable. The coefficients can be positive, negative, or close to zero depending on the sample that is chosen. We discuss the implications of this finding for monetary policy.

6. Conclusions

This paper has revisited the question of whether oil price volatility is useful as a predictor of stock price volatility. There is a strong, positive contemporaneous relationship between the two volatility series. Consistent with previous studies, there is clear evidence of a predictive relationship when doing inference on the full sample.

The results are different when we evaluate of the ability of oil price volatility to improve out-of-sample forecasts of stock return volatility. Formal out-of-sample predictive ability tests find no evidence that oil price volatility can be used to improve forecasts of stock return volatility.11 Further investigation reveals thatthe relationship between the two volatility series fluctuates wildly through time.12 The changes in the relationship are not just in magnitude, but also in sign. One could find a strong positive relationship, a strong negative relationship, or no relationship at all, simply by choosing an appropriate subsample. Therefore, in spite of the reasonableness of the argument that there should be a link between stock market volatility and oil market volatility, we conclude that it cannot be exploited in practice.

Our results provide no support for the hypothesis that oil price volatility should be used as a predictor of stock return volatility. Thus, monetary and fiscal policy authorities should not adjust policy in response to high oil price volatility, unless there are other concerns about oil price volatility beyond the effects on stock price volatility.