A Hybrid Approach to Portfolio Optimization Using Technical Analysis and Data Mining

Investing in the stock market, is a significant part of the country's economy. Increasing profits and reducing the risk of investing in the stock exchange has always been a major concern for investors. Also, Stock markets are affected not only by macroeconomic parameters but also by thousands of other factors. This research aims to provide a model in which future stock potential is forecasted by considering the technical analysis indicators by the fuzzy neural network. According to the forecasts, the mathematical model based on factors such as the return, variance, and skewness of the stock portfolio will be optimized. Then, this model is solved using the genetic algorithm. This research is an applied research in terms of purpose and is descriptive in terms of method. The empirical results shows that the proposed models will provide more profit to investors regarding variance and skewness comparing to traditional models and stock market index.