### State-value weighted entropy as a measure of investment risk.

An examination is made of the use of the entropy measure as a measure of investment performance. The commonly used entropy measure disregards the dispersion of security frequency classes used in the calculation of entropy. Thus, state-value weighting of the entropy is proposed and tested using a portfolio selection heuristic algorithm. Nawrocki's (1983) methodology is replicated, and the sample consists of 62 securities randomly selected from the CRSP tape, with data covering the period 1950-1975. The results suggest that weighting the entropy value will increase the investment performance of the entropy risk measure. Two alternative calculations for entropy are suggested: 1. entropy weighted by squared deviations from the mean, and 2. entropy weighted by absolute deviations from the mean. Other undominated portfolios are the target semivariance estimate and the Elton, Gruber and Padberg (1976) optimal algorithm.

Main Author: Nawrocki, David. Harding, William. Villanova Faculty Authorship English 1986 http://ezproxy.villanova.edu/login?url=https://digital.library.villanova.edu/Item/vudl:178261
Summary: An examination is made of the use of the entropy measure as a measure of investment performance. The commonly used entropy measure disregards the dispersion of security frequency classes used in the calculation of entropy. Thus, state-value weighting of the entropy is proposed and tested using a portfolio selection heuristic algorithm. Nawrocki's (1983) methodology is replicated, and the sample consists of 62 securities randomly selected from the CRSP tape, with data covering the period 1950-1975. The results suggest that weighting the entropy value will increase the investment performance of the entropy risk measure. Two alternative calculations for entropy are suggested: 1. entropy weighted by squared deviations from the mean, and 2. entropy weighted by absolute deviations from the mean. Other undominated portfolios are the target semivariance estimate and the Elton, Gruber and Padberg (1976) optimal algorithm.