Abstarct: Advances in artificial intelligence and machine learning, especially in evolutionary computing, have not only enabled us to analyze data more effectively, but have also enabled them to be used to understand any underlying pattern of financial markets. The use of methods to predict the future has always been the main concern of thinkers in various sciences. In this way, naturally, they have methods, durability and suitable application that have the least possible error in forecasting. The Gray Wolf Algorithm is a computer science search technique for finding approximate solutions to optimization and search problems, which is a special type of evolutionary algorithm and is simulated from biological techniques such as gray wolf hunting. Predicting stock prices in investments is of particular importance because it is considered an important factor in stock valuation methods and in most cases is the main factor in stock investment decisions. In the market, what is more important for the participants than predicting the exact rate of growth of a share price is the issue of buying or selling, that is, a great growth can lead to a decision to buy, and vice versa if predicting a sharp decline with high probability It can be a decision to sell a share. The important issue here is to determine these different phases or scenarios that must be done accurately by experts and market participants. One of the most important factors is determining the amount of stock bought or sold. In fact, in this case, the user determines the amount of stock sales according to the distance between the price of the last transaction and the final price. So the amount of sales has a huge impact on the amount of profit received from stock trading, so that its optimal determination can lead to profit maximization. In this dissertation, in order to search for a successful strategy for trading in the Iranian stock market, an approach baxsed on the Gray Wolf algorithm is presented. The required data have been collected from the Iranian Stock Exchange Organization from 1394 to 1399 for 10 active companies. The proposed method, given the price of the last trade as well as the final price, tries to present a strategy, which tells the trader when to enter and leave a trade. It is further shown that the proposed approach has been able to increase the size of the data set, the number of people and the number of iterations by 52% of traders.