Stock prices are influenced by economic, technological, psychological and geopolitical factors. A review of the literature in this field shows that stochastic approaches, trend analysis and econometrics have been used to demonstrate stock market dynamics and price forecasting. However, these techniques cannot provide a comprehensive overview of market dynamics. Because they ignore the temporal relationship between these factors and are unable to understand their cumulative effects on prices. By integrating chaos theory and continuous data mining based on price acceleration, this study has eliminated these gaps by inventing a new price forecasting method called dynamic stock market recognition simulator and combining two methods: one is delay structures. Or gives time intervals to the data set, and the other is the method of selecting new variables for the market environment. The results showed that the method used can be used to predict the long-term stock price using a small data set with small dimensions.
