Two essays in finance and statistics: Building tracking portfolios and designing risk management programs

This dissertation contains two essays. The first essay concerns how to build a tracking portfolio of stocks whose return of investment mimics that of a chosen investment target. Statistically, this task can be accomplished by selecting an optimal model from constrained linear models. To develop an automatic procedure for building an optimal tracking portfolio, we extend the Generalized Information Criterion (GIC) to constrained linear models either with independently and identically distributed random errors or with dependent errors that follow a stationary Gaussian process. The asymptotic validity of the extended GIC is established. Simulation results show that the relative frequency of selecting the optimal constrained linear model by the GIC is close to one in finite samples. We apply the GIC based procedure for building an optimal tracking portfolio to the problem of measuring the long-term impact of a corporate event on stock returns and demonstrate empirically that it outperforms two other competing methods.;The second essay concerns how corporations organize their risk management program. We set up a theoretical framework to analyze the problem of designing a risk management program for multidivisional corporations. Our analysis shows that risk management programs currently existing in large corporations are not optimal. We propose a new risk management program in which the corporate headquarters organizes an internal market for divisions to trade state-contingent claims among themselves. We show that this new program is better than existing ones.