Om Parkash and Mukesh
Keywords: Portfolio selection theory, cross entropy, mean-variance efficient frontier, uncertainty
Abstract: It is a well-known fact that in the literature of information theory, a variety of divergence (distance or cross entropy) measures is available, each with its own merits and limitations. These measures are applicable to various disciplines of Mathematical Sciences. One such discipline pertaining to Operations Research is portfolio analysis. In the present communication, we have developed two new parametric measures of cross entropy and consequently provided the applications of these measures for the study of optimization principles for the development of measures of risk in portfolio analysis. We have observed that minimizing these measures implies the minimization of the expected utility of the risk-prone person and maximization of the expected utility of a risk-averse person.
[View Complete Article]