Working Papers
Oil Jump Risk (Job Market Paper), Forthcoming, Journal of Futures Markets
We show that the risk premium associated with the big upward jumps in the oil markets is a significant driver of the cross-section of stock returns, while the other two premia are not significant. The results show that the highest significance is for the 2000-2008 period, which is called the "commodity super-cycle'' period in the literature of commodity-finance, a period during which the demand of oil was growing considerably and the supply was not keeping up with the demand because of the low spare capacity of main oil producers in the world. The results unfold that after the emergence of the "Shale Revolution'' in the 2009-2011 period, the significance of the premium as a driver of the cross-section of stock returns is less than the previous period. Our explanation for this observation is that the technological revolution has made US companies able to increase their production considerably, which makes the US economy less vulnerable (than the recent decades) to oil import, and as a result to the shocks from oil prices which were mainly driven by geopolitical incidents in the middle-east. We also show that the risk premia of big downward and upward jumps in oil prices have a considerable prediction power for important economic indicators such as gross domestic product growth, consumption growth, and total investment growth. Moreover, we see that the upside jump premium is a significant and relatively strong predictor for fundamentals of oil markets like inventory growth, demand growth, and OPEC’s production growth.
The Risks of Skewness and Kurtosis in Oil Market and the Cross-Section of Stock Returns (with Craig Pirrong)
We show that exposure to the risk of kurtosis in oil market drives the cross-section of stock returns from 1996 to 2014. The average monthly difference between the return of portfolio of stocks with low exposure and high exposure to the risk of kurtosis is -0.37%, showing that higher exposure to oil’s kurtosis risk will be penalized by lower average returns. We are able to confirm the significance of kurtosis risk within the statistical framework of Carhart 4-factor model. In contrast to the skewness risk, which is only a significant player in some of the sub-periods, kurtosis risk is keeping its significance through all sub-periods, as well as after taking market moments into account and within different maturities. The significance of the risk of skewness gets more evident moving from shorter to longer maturities. The risk of volatility, which has been shown to be a significant-priced risk in the cross-section of stock returns in literature, loses its significance after controlling for the third and fourth moments.
The Fundamental, Speculative and Macroeconomic Determinants of Variance and Skew Risk Premia in Oil Market (with Craig Pirrong)
Using the model-independent approaches of Trolle and Schwartz (2008) and Kozhan et al (2013), we estimate the Variance Risk Premium and Skew Risk Premium for oil market. After estimation, the contribution of the paper is twofold. First, we try to figure out which variables can describe the variation in variance and skew risk premium and the risk-neutral skewness. The results show that we are able to describe a considerable portion of the variation in all three variables using speculative and macroeconomic factors and most of the variation is being described by by the latter group of variables. We also try to capture the effect of fundamentals of oil market on these variables. The results show that the fundamentals are able to describe big portion of variation in these three variables. Second, we try to predict the return of delta-hedged and delta-vega hedged option portfolios and also oil futures return. The results show that variance risk premium is a significant predictor of delta-hedged portfolio cumulative return with a positive sign while skew risk premium can describe the cumulative return of the delta-Vega Hedged portfolio with negative sign. The results also show that both variance and skew risk premia are significant predictors of cumulative futures return with different signs.
Oil Jump Risk (Job Market Paper), Forthcoming, Journal of Futures Markets
We show that the risk premium associated with the big upward jumps in the oil markets is a significant driver of the cross-section of stock returns, while the other two premia are not significant. The results show that the highest significance is for the 2000-2008 period, which is called the "commodity super-cycle'' period in the literature of commodity-finance, a period during which the demand of oil was growing considerably and the supply was not keeping up with the demand because of the low spare capacity of main oil producers in the world. The results unfold that after the emergence of the "Shale Revolution'' in the 2009-2011 period, the significance of the premium as a driver of the cross-section of stock returns is less than the previous period. Our explanation for this observation is that the technological revolution has made US companies able to increase their production considerably, which makes the US economy less vulnerable (than the recent decades) to oil import, and as a result to the shocks from oil prices which were mainly driven by geopolitical incidents in the middle-east. We also show that the risk premia of big downward and upward jumps in oil prices have a considerable prediction power for important economic indicators such as gross domestic product growth, consumption growth, and total investment growth. Moreover, we see that the upside jump premium is a significant and relatively strong predictor for fundamentals of oil markets like inventory growth, demand growth, and OPEC’s production growth.
The Risks of Skewness and Kurtosis in Oil Market and the Cross-Section of Stock Returns (with Craig Pirrong)
We show that exposure to the risk of kurtosis in oil market drives the cross-section of stock returns from 1996 to 2014. The average monthly difference between the return of portfolio of stocks with low exposure and high exposure to the risk of kurtosis is -0.37%, showing that higher exposure to oil’s kurtosis risk will be penalized by lower average returns. We are able to confirm the significance of kurtosis risk within the statistical framework of Carhart 4-factor model. In contrast to the skewness risk, which is only a significant player in some of the sub-periods, kurtosis risk is keeping its significance through all sub-periods, as well as after taking market moments into account and within different maturities. The significance of the risk of skewness gets more evident moving from shorter to longer maturities. The risk of volatility, which has been shown to be a significant-priced risk in the cross-section of stock returns in literature, loses its significance after controlling for the third and fourth moments.
The Fundamental, Speculative and Macroeconomic Determinants of Variance and Skew Risk Premia in Oil Market (with Craig Pirrong)
Using the model-independent approaches of Trolle and Schwartz (2008) and Kozhan et al (2013), we estimate the Variance Risk Premium and Skew Risk Premium for oil market. After estimation, the contribution of the paper is twofold. First, we try to figure out which variables can describe the variation in variance and skew risk premium and the risk-neutral skewness. The results show that we are able to describe a considerable portion of the variation in all three variables using speculative and macroeconomic factors and most of the variation is being described by by the latter group of variables. We also try to capture the effect of fundamentals of oil market on these variables. The results show that the fundamentals are able to describe big portion of variation in these three variables. Second, we try to predict the return of delta-hedged and delta-vega hedged option portfolios and also oil futures return. The results show that variance risk premium is a significant predictor of delta-hedged portfolio cumulative return with a positive sign while skew risk premium can describe the cumulative return of the delta-Vega Hedged portfolio with negative sign. The results also show that both variance and skew risk premia are significant predictors of cumulative futures return with different signs.
