My dissertation focuses on two areas of financial economics. First, I look at the purchase and sale of operating units by companies. The first chapter characterizes the behavior of value-maximizing firms, which may invest in new capital, purchase existing assets or sell assets. This approach yields an endogenous selection model that links asset purchases and sales to fundamental properties of the firm. Empirical tests confirm the predictions of the model. In particular, return on assets and size strongly predict when firms purchase or sell assets, and the size of the transaction covaries with the marginal value of capital. The second chapter explores cross-sectional variation in abnormal returns associated with asset purchases and sales. Ceteris paribus , a dollar increase in the transaction size leads to a 11¢ gain to buyers and a 14¢ gain to sellers. This demonstrates that asset sales improve the allocative efficiency of capital. However, this is tempered by agency problems. The market discounts asset purchases by firms with poor governance measures, and acquisitions by large firms lead to lower returns. Further, while a dollar increase in transaction size creates 21¢ of value to strong governance firms, the corresponding value for weak governance firms equals only 3¢. The third chapter explores return predictability from the perspective of a skeptical investor. Previous literature has tended toward two polar viewpoints: predictability is useful only if the statistical evidence is incontrovertible, or that predictability should affect portfolio choices even if the evidence is weak. This paper models an intermediate view. We investigate optimal portfolio choice for an investor who is skeptical about the amount of predictability in the data. Skepticism is modeled as an informative prior over the improvement in the Sharpe ratio generated by using the predictor variable. The resulting weights are less volatile, and, as we show, deliver superior out-of-sample performance compared with weights implied by diffuse priors, dogmatic priors, and regression analysis.