A model for clustering and optimizing portfolio: Tehran stock exchange using data mining algorithms

Management of investment basket and selecting assets is one of the problems of decision making in financial area. In the competitive business environment, in order to confront complex competitions in the market, financial institutes try to consider the best policy of investment basket that in turn leads to an increase in the output for the investors. The goal of this study is to develop a portfolio by considering the behavior of investors in risk taking in a realistic method. This research aims at supporting investors, experts and intermediate managers in establishing optimized portfolio of stocks. The proposed model has used the data of 66 stockholders who were enlisted in Stock Exchange Market by using the five indexes of risk output, skewness, liquidity and current ratio and clustered different companies by using the neuro-networks SOM algorithm The results show that the function of model to general index, the industry index and the index of 50 more active companies are better in Tehran Stock Exchange.