Asset pricing with Levy jump processes

This thesis comprises of three essays that explore the theoretical development as well as the empirical applications of asset pricing models with Levy jump processes. The first essay presents a new discrete-time framework that combines heteroskedastic processes with rich specifications of jumps in returns and volatility. Our models can be estimated with ease using standard maximum likelihood techniques. We evaluate the models by fitting a long sample of S&P500 index returns, and by valuing a large sample of options. We find strong empirical support for time-varying jump intensities. A model with jump intensity that is affine in the conditional variance performs particularly well both in return fitting and option valuation. In the second essay, we develop a new class of asset pricing model that combines the flexibility of Levy processes with the ease of implementation of affine GARCH dynamics. This framework produces a large class of asset return processes that have analytical solutions for their pricing transform, and lead to a simple valuation of derivatives. We apply this newly proposed framework to various two-factor models consisting of a normal and a pure jump Levy component. The results from joint estimation of options and returns on the market index reveal the important economic role of jumps. We find that models without jumps cannot reconcile the difference between market-realized returns and investors' ex-ante expectations of returns with an economically justifiable equity premium level. In the third essay, we provide evidence that the market crash risk is priced in individual equity options. We proceed by developing a factor model for equity returns and option pricing that takes into account the market's systematic risk factors, namely the market volatility and jump risks. The probability of large negative jumps in the market return produces the "crash fear" effect. In addition to the market risk factors, we allow each firm to be affected by its own equity-specific shock. The estimation results show that the market jump and volatility risk factors are priced and explain the differential price structures among individual equity options. We find that in the cross section, the average compensation for bearing the market jump risk is 3.18 percent in terms of annualized excess return.