Modeling and Estimating the return of Tehran Stock Exchange using dynamic models

Since the creation of the stock market in the nineteenth century, many researchers have focused on research into stock price forecasting models and market returns. Statistical prediction models such as Arma, Arima, Arch, have been widely used but none of them have had the desired result. Therefore, many researchers have recently considered the stock market as a nonlinear dynamic system. The application of nonlinear models as well as advanced techniques, although not many years have begun, but in a short time has been able to open its place in various sciences. The purpose of this study is to predict the stock index using the dynamic model averaging DMA and also the method of the dynamic model selective DMS and the use of quarterly data for the years 1380-1399. The main advantage of the model used in the present study is the introduction of a large number of independent variables for its dynamics without the usual problem of overfitting appearing in the model. In this paper, the effect of some macroeconomic variables on the process of modeling and forecasting stock returns on the stock exchange was investigated. The results of the article showed that the probability of entering the variables of money supply growth, quasi-money growth, inflation, land price index growth in large cities is more than other input variables.