Abstract<\/strong><\/p>\n The optimal portfolio selection problem has always been the most important issue in the modern economy.\u00a0 In this Study, It has been shown that how an investment with n risky share can achieve the certain profits with less risk that spread between stocks. Such a portfolio, it is called an efficient portfolio and it is necessary to find solving the optimization problem. Hence, the Improved Particle Swarm Optimization algorithm is used. The value of Portfolio and its risk are applied as the parameters in optimizing aim and criterion value exposed to contingent risk. Three intended applications have been indicated to the portfolio. In the next stage ,to evaluate and validate the method and to estimate the value of the portfolio in the next days and hold the series of the stock prices ,within a specified period, to predict the price and The Autoregressive method algorithms is used for modeling of the time-series. Practical result achieved for solving the portfolio optimization problem in Tehran Stock Exchange for the next day, by choosing the basket which includes 20 companies among the 30 most active industry indicates the performance and high capability of the algorithms and used in solving constrained optimization and appropriate weighted portfolios.<\/p>\n