Designing a Model for Using Artificial Neural Networks to Predict Nonlinear time Series (Case study: Tehran Stock Exchange Index)
Subject Areas : Futurology
Keywords: Total Stock Index, Prediction, Artificial Neural Network, Tehran Stock Exchange,
Abstract :
Introduction: Forecasting the total stock index is a challenging task, due to the complexity of stock market variables and the lack of problem management in critical times, it is very difficult to develop an efficient model for forecasting the total stock index. Relatively accurate prediction of index movement is very important and vital for capital market investors. One of the important tools used for investment decisions is forecasting techniques, which are an integral part of the decision-making and control process. On the other hand, forecast accuracy has a direct relationship with decision risk. This means that the more accurate the forecast, the lower the loss or risk from decision-making under uncertainty. One of the well-known and new methods for predicting the total stock index is the method of using artificial neural networks. Purpose: The main purpose of this research is to present the optimal model of using artificial neural networks to predict non-linear time series (case study: Tehran Stock Exchange Index) and this research is practical in terms of purpose. Research method: In terms of research method, it is descriptive based on survey and in terms of review method, analytical-mathematical. The statistical population of this research is the index of the entire Tehran Stock Exchange from 1369 to 1399, and in this research, the measurement tools and variables are the documents and statistics of the Tehran Stock Exchange, and the data analysis in this research is based on statistical methods. Descriptive and inferential statistics as well as artificial neural networks have been used. Perceptron layer is used. Findings: The results of this research confirm the high accuracy of predicting the total index of Tehran Stock Exchange compared to other estimation methods using the presented model, which has the ability to predict the total index with an error of 1.4%. Conclusion: Confirming that the Tehran Stock Exchange index follows a non-linear process is considered one of the main and important results of this research, and at the end, practical suggestions for users and researchers in future researches are presented.