This dissertation consists of two essays on crude oil futures and options markets. The first essay investigates whether aggregate risk aversion and risk premiums in the crude oil market co-vary with the level of speculation. Using crude oil futures and option data, I estimate aggregate risk aversion in the crude oil market and find that it is signi ficantly lower after 2002, when speculative activity started to increase. Using speculation index as a state variable, risk premiums implied by the state-dependent risk aversion estimates confi rm the negative correlation between speculative activity and risk premiums, and indicate that risk premiums in the crude oil market are on average lower and more volatile after 2002. These findings suggest that index-fund investors who demand commodity futures for the purpose of portfolio diversi fication are willing to accept lower compensation for their positions. Estimated state-dependent risk premiums have substantial predictive power for subsequent futures returns and outperform commonly used predictors. The second essay exams the economic importance of jumps, jump risk premiums, and dynamic jump intensities in crude oil futures and options markets. Existing pricing models for crude oil options are computationally intensive due to the presence of latent state variables. Using a panel data of crude oil futures and options, I implement a class of computationally e fficient discrete-time jump models. I find that jumps account for about half of the total variance in crude oil futures and options prices, and a substantial part of the risk premiums is due to jumps. Jumps are large and rare events in crude oil futures and options markets. The main role of jumps and jump risk premiums in crude oil futures and options markets is to capture excess kurtosis in the data. These findings suggest that it is critical to include jumps in pricing models for crude oil futures and options, and there is strong evidence in favor of time-varying jump intensities.

"This thesis consists of two essays on the pricing of equity and commodity derivatives. In the first essay, we investigate overpricing of call options. A recent empirical study by Constantinides, Czerwonko, Jackwerth and Perrakis (2011) documents that S&P 500 Index call options are frequently overpriced in the sense that any rational agent can improve her expected utility by writing these calls. Little work has addressed why such apparent mispricing is so common. We show that such overpricing of call options is consistent with equilibrium in an economy where investors have portfolio constraints and heterogeneity in beliefs on both expected return and volatility. Within our model, call options are overpriced when belief dispersion is large and the capacity of liquidity providers is small. Empirically, we propose a model-free methodology to investigate the determinants of option overpricing and verify my explanation. The findings are robust to various implementations of the empirical study.In the second essay, we study the variance risk premia calculated from the crude oil futures and options market in a model-free way. First, we establish that the variance risk premia are negative for various maturities, a finding that reflects the compensation for crude oil option writers. While the existing literature focuses on one month maturity, we analyze maturities beyond one month because commodity hedging demands for futures and options often have longer horizons. Furthermore, for the first time in the literature, we document that the variance risk premia predict the commodity futures returns after several information variables, such as storage level and hedging pressure, are controlled for. The finding is robust across various implementations of predictive regressions and out-of-sample tests. Finally, we develop a stylized economic model to show that the hedge demand for both futures and options may explain such predictability." --

This thesis consists of three essays on risk management in crude oil markets. In the first essay, the valuation of an oil sands project is studied using real options approach. Oil sands production consumes substantial amount of natural gas during extracting and upgrading. Natural gas prices are known to be stochastic and highly volatile which introduces a risk factor that needs to be taken into account. The essay studies the impact of this risk factor on the value of an oil sands project and its optimal operation. The essay takes into account the co-movement between crude oil and natural gas markets and, accordingly, proposes two models: one incorporates a long-run link between the two markets while the other has no such link. The valuation problem is solved using the Least Square Monte Carlo (LSMC) method proposed by Longstaff and Schwartz (2001) for valuing American options. The valuation results show that incorporating a long-run relationship between the two markets is a very crucial decision in the value of the project and in its optimal operation. The essay shows that ignoring this long-run relationship makes the optimal policy sensitive to the dynamics of natural gas prices. On the other hand, incorporating this long-run relationship makes the dynamics of natural gas price process have a very low impact on valuation and the optimal operating policy. In the second essay, the relationship between the slope of the futures term structure, or the forward curve, and volatility in the crude oil market is investigated using a measure of the slope based on principal component analysis (PCA). The essay begins by reviewing the main theories of the relation between spot and futures prices and considering the implication of each theory on the relation between the slope of the forward curve and volatility. The diagonal VECH model of Bollerslev et al. (1988) was used to analyse the relationship between of the forward curve slope and the variances of the spot and futures prices and the covariance between them. The results show that there is a significant quadratic relationship and that exploiting this relation improves the hedging performance using futures contracts. The third essay attempts to model the spot price process of crude oil using the notion of convenience yield in a regime switching framework. Unlike the existing studies, which assume the convenience yield to have either a constant value or to have a stochastic behaviour with mean reversion to one equilibrium level, the model of this essay extends the Brennan and Schwartz (1985) model to allows for regime switching in the convenience yield along with the other parameters. In the essay, a closed form solution for the futures price is derived. The parameters are estimated using an extension to the Kalman filter proposed by Kim (1994). The regime switching one-factor model of this study does a reasonable job and the transitional probabilities play an important role in shaping the futures term structure implied by the model.

Trading in oil futures and options is an introduction to price risk management in the worldwide oil industry. With numerous practical examples, it requires no prior knowledge and should be read by everyone involved in the industry. Although aimed primarily at those new to risk management it will also provide a useful theoretical background to more experienced managers and it will show those in other markets how the oil industry uses futures and other derivatives. This book concentrates on all the risk management tools available to everyone from crude oil producer to refined product consumer and explains the theory of futures, exchange options and over the counter trading.

As OPEC has loosened its grip over the past ten years, the oil market has been rocked by wild price swings, the likes of which haven't been seen for eight decades. Crafting an engrossing journey from the gushing Pennsylvania oil fields of the 1860s to today's fraught and fractious Middle East, Crude Volatility explains how past periods of stability and volatility in oil prices help us understand the new boom-bust era. Oil's notorious volatility has always been considered a scourge afflicting not only the oil industry but also the broader economy and geopolitical landscape; Robert McNally makes sense of how oil became so central to our world and why it is subject to such extreme price fluctuations. Tracing a history marked by conflict, intrigue, and extreme uncertainty, McNally shows how—even from the oil industry's first years—wild and harmful price volatility prompted industry leaders and officials to undertake extraordinary efforts to stabilize oil prices by controlling production. Herculean market interventions—first, by Rockefeller's Standard Oil, then, by U.S. state regulators in partnership with major international oil companies, and, finally, by OPEC—succeeded to varying degrees in taming the beast. McNally, a veteran oil market and policy expert, explains the consequences of the ebbing of OPEC's power, debunking myths and offering recommendations—including mistakes to avoid—as we confront the unwelcome return of boom and bust oil prices.

This dissertation studies two major topics related to the crude oil price and the economy. The first topic studied is about the relationship between speculation and the crude oil price and the related implications on the macroeconomic growth and inflation. The second topic is about the relationship between the oil price volatility and the US stock market. It includes two subtopics: i) the volatility spillovers between the crude oil market and the US stock market and ii) the relationship between oil price volatility and real stock returns on the US market. This dissertation has four chapters, with each of the two major topics studied relatively independently in their respective chapters. In the first chapter, we introduce the background and motivation for the topics studied in this dissertation. Additionally, we also give an overview of the results and important findings. In the second chapter, we examine the impact of speculative information on the oil price and the corresponding implications on the macroeconomy. We use a structural vector autoregression (VAR) model to decompose the shocks of the crude oil price and use the gold price as a proxy for the speculative information. We argue that using the gold price to account for speculative information is a very informative alternative to the other indicators used in literature. Our results show that speculative information plays a very important role in driving crude oil price shocks; it accounts for about 20% of the variation of the oil price. Furthermore, we show that speculative shocks to the crude oil price are correlated to future macroeconomic downturns. We also show that speculative shocks may create inflation pressure, although the effect is not as strong as that on the macroeconomic output growth. In the third chapter, we use a generalized autoregressive conditional heteroskedasticity (GARCH) specification to model the volatility on both the oil and stock markets and then utilize an extension of the GARCH-M (GARCH in mean) vector autoregression (VAR) model introduced in Elder (2004) to capture the volatility spillover relationship between the two markets and the relationship between the volatility of the oil price and stock returns at the same time. Further, we detect a structural change of the oil price-stock returns relationship near the middle of 1987. A unidirectional volatility spillover from the stock market to the oil market is found to be statistically significant before the break, while a negative relationship between oil price volatility and the conditional mean of stock returns is more pronounced afterwards. We argue that several events happening around the break point are likely to be the causes for the structural change. In the last chapter, we summarize the work and highlight the important results in this dissertation. In addition, we also discuss possible future research directions.