UC Berkeley

This dissertation addresses various aspects of asset pricing theory in the following three contexts: the case of insider trading (of stocks) with uninformed biased traders, the case of trading of real options (specifically, of the option to sell a real indivisible asset), and the case of house pricing and construction of better house price indices.

Chapter 1 examines the effects of insider trading on uninformed traders with bounded rationality in the context of a continuous-time Kyle-type model with a single perfectly informed risk-neutral agent (insider), a competitive risk-neutral market maker and a set of biased uninformed traders.

Two cases of behavioral biases or bounded rationality on the part of the uninformed traders are considered. In the first case the uninformed traders' order flow has a non-zero covariation with a set of public signals (where positive covariation describes aggregate momentum strategies among the uninformed investors in reaction to news, while negative covariation indicates that the uninformed traders are predominantly contrarians). In the second case, the order flow from the uninformed traders has a strictly positive or a strictly negative covariance between its increments and is no longer Markov.

The equilibrium strategy of the insider, taking into account such biases, is derived in both cases and the

effects of the biases on the equilibrium price of the underlying asset are considered. The question of whether such biases benefit or harm the uninformed traders is answered.

In Chapter 2 a class of mixed stochastic control/optimal stopping problems arising in the problem of finding the best time to sell an indivisible real asset, owned by a risk averse utility maximizing

agent, is considered. The agent has power type utility based on the $\ell_{\alpha}$-type aggregator and has access to a frictionless financial market which can be used to partially hedge the risk associated with the real asset if correlations between the financial assets and the real asset value are nonzero. The solution to the problem of finding the optimal time to sell the real asset is characterized in terms of solution to a certain free boundary problem. The latter involves a nonlinear partial differential

equation and includes, as special case with $\alpha=1$, the Hamilton-Jacobi-Bellman equation found in {Evans, Henderson, Hobson, 2008}. Comparisons with the case of exponential utility are also given.

Due to lack of data, the U.S. primarily uses repeat-sales indices to measure real-estate returns, despite the serious shortcomings of these indices. Making use of a newly available data set that contains both time-varying characteristics for all properties in the U.S. and transaction details for those properties that traded, in Chapter 3 a new hedonic house-price index is developed that overcomes these shortcomings by allowing house prices and returns to depend on house characteristics and on local and national macroeconomic factors. The index is estimated using Markov Chain Monte Carlo (MCMC) linear filtering techniques and results in significant differences, in both the level and volatility of prices, between the new estimates and those from the Federal Housing Finance Board's weighted-repeat-sales (WRS) price index. This suggests that the new index is significantly superior to repeat-sales indices as a measure of U.S. real-estate returns for economic forecasting, mortgage valuation, and bank stress tests.