A genetic network programming model for portfolio optimization by generating risk-adjusted trading rules

Genetic network programming (GNP) as an evolutionary computation method has been used for stock trading recently. Former researches confirm the efficiency of trading rules which are created by GNP. In this paper, GNP has been applied for stock portfolio optimization by generating risk-adjusted trading rules. There are two main novelties in this paper: 1) we use conditional Sharp ratio as a risk-adjusted measure for generating trading rules, 2) in our GNP model, binary trading rules have been extended to more realistic rules which are called trinary rules using three signals of buy, sell and no trade. We applied our GNPmodel on ten stocks from Tehran Stock Exchange (TSE). The numerical results show that our proposed model with three signals outperformed the previous model with two signals of buy and sell in terms of excess return and excess risk adjusted return.