An Optimal Stock Portfolio Construction Model Using Genetic Algorithm

Investing in share market is a risky business. While some of the risks can be controlled, others can only be guarded against. To reduce the risk of an investment, many portfolio selection methods have been proposed. To generate an optimal stock portfolio, one has to select stocks and decide the proportion of the capital to be invested on each stock. In this paper, we present a genetic algorithm to generate stock portfolios. The proposed model uses the concept of graph theory to predict the prices of stock which are used as an input to the fitness function of the genetic algorithm to generate optimal portfolio. The proposed model gave good results when experimented on stocks of Bombay Stock Exchange.