The classification of parametric choices under uncertainty: Analysis of the portfolio choice problem

This paper describes the admissible classes of parametric distribution functions of return portfolios and analyzes their consistency with the maximization of the expected utility. In particular, we present a general theory and a unifying framework with the following aims: (1) studying the implications of the classical market restrictions on the portfolio distributions; (2) establishing general rules of ordering, when the uncertain prospect depends by a finite number of parameters; (3) understanding how a dispersion measure has to be used, in order to obtain the investors' optimal portfolios.