| Most empirical studies of the static CAPM assume that betas remain constant over time and that the return on the value-weighted portfolio of all stocks is a proxy for the return on aggregate wealth. The general consensus is that the static CAPM is unable to explain satisfactorily the cross-section of average returns on stocks. In this thesis, it is assumed that the CAPM holds in a conditional sense, i.e., betas and the market risk premium vary over time. The return on human capital is included when measuring the return on aggregate wealth. The model specified in this essay performs well in explaining the cross-section of average returns.; The thesis also contributes in two areas to the econometric methods for finance. First, it provides a formula for correcting the bias in the widely used two-pass regression method. Second, it derives the asymptotic distribution of the minimized objective function in the Generalize Method of Moments with an arbitrary weighting matrix, which may be of broad interest to economists.; Although the existing empirical studies for portfolio efficiency assume away constraints on portfolio positions, investors often encounter stringent restrictions on negative holdings. In the mean-variance framework, the market portfolio may be inefficient if asset holdings must be non-negative, but it is efficient if there exists a risk-free borrowing and lending rate.; A general Bayesian approach is introduced in this thesis to examine the degree of portfolio inefficiency when investors face constraints on portfolio weights. Using this approach, the NYSE-AMEX market portfolio is found inefficient when portfolio weights are constrained to be non-negative, but the degree of inefficiency is much smaller than when the weights are unconstrained. Constraints on negative holdings also sharply reduce uncertainty in the inference on portfolio efficiency. These results are robust to deviations from the assumption that investors possess full information. |
