This study, given its Bayesian approach, is related to the recent article by Baks, Metrick, and Wachter (2001), who estimate funds' alphas using informative prior beliefs about alpha. They investigate the degree to which informative priors can preclude an investor from infer- ring that at least one actively managed fund has a positive alpha. This inference relates to an investment problem of a mutual fund investor who can also earn the hypothetical costless returns on the benchmark indexes. In that setting, if a given fund's alpha is greater than zero, then combining that fund with a position in the benchmarks produces a higher Sharpe ratio than an.
