Mainak Dey¹ and Rupak Bhattacharyya², ¹Camellia Institute of Engineering, India and ²Global Institute of Management & Technology, India

This paper introduces a new stock portfolio selection model in non-stochastic environment. Following the principle of maximum entropy, a new entropy-cost ratio function is introduced as the objective function. The uncertain returns, risks and dividends of the securities are considered as interval numbers. Along with the objective function, eight different types of constraints are used in the model to convert it into a pragmatic one. Three different models have been proposed by defining the future financial market optimistically, pessimistically and in the combined form to model the portfolio selection problem. To illustrate the effectiveness and tractability of the proposed models, these are tested on a set of data from Bombay Stock Exchange (BSE). The solution has been done by genetic algorithm.

Stock Portfolio Selection, Interval Numbers, Entropy-Cost Ratio model, Dividend, Genetic Algorithm.