| In the first essay, we investigate the temporal pattern of stock prices in an equilibrium that aggregates the demand functions of both rational investors and agents who sell their winners too soon but hold on to their losers for too long. Even when a stock's fundamental value follows a random walk, its equilibrium price will underreact to information and becomes predictable by its unrealized capital gains. Stocks on which most investors experienced capital gains have higher expected returns than those that have experienced large price declines. We prove analytically that a momentum strategy is profitable. Empirically, we find that when the capital gains is used as a regressor along with past returns and volume to predict future returns, the momentum effect disappears.; In the second essay, I develop a string model that explicitly incorporates stochastic volatilities and correlations of bond yields, and derive closed-form approximation for the model prices of European interest rate caps and swaptions. The model is econometrically estimated with a panel dataset on swap rates and swaptions. Empirically, one- and two-factor model are rejected both statistically and economically in favor of our three-factor model, which explains the time-series and cross-sectional variations in swaption implied volatilities very well and eliminates most of the large mis-pricing between swaptions and caps found in previous studies. The implied variances of bond yields explain almost all of the realized variances, while the GARCH estimates have no incremental forecasting power.; The third essay provides new evidence on the relation between firm performance and insider ownership using a panel data of real estate investment trusts (REITs). Consistent with the, trade-off between incentive alignment and entrenchment effect of managerial equity ownership, I find a significant roof-shaped relation between Tobin's Q and insider ownership. About 60% of the REITs in my sample are Umbrella Partnership REITs that suffer from a unique form of agency cost arising from differing tax basis between limited partners and common shareholders. At any given level of insider ownership, Tobin's Q of an UPREIT is on average lower than a regular REIT, other things equal. The optimal level of insider ownership for UPREITs is also significantly lower than regular REITs. |
