Prediction of Food Industry Price Index Using Artificial Neural Network (Case Study: Index of Tehran Stock Exchange)

Abstract

Forecasting the total stock index is a challenging task, an accurate forecast of the movement of the index is very important for the performance of investors, due to the complexity of the stock market the lack of management, and the occurrence of problems in critical times, it is challenging to develop a useful model for forecasting. One of the important tools used for investment decisions is forecasting techniques, which is an integral part of the decision-making and control process, on the other hand, it has a direct relationship with decision-making risk. This means that the more accurate the prediction is, the loss or risk caused by decision-making in conditions of uncertainty is reduced. One of the well-known and new methods of predicting the total stock index is the artificial neural network (ANN) method. The main goal 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). This research is descriptive based on the survey and is analytical-mathematical in terms of the survey method. The statistical population of this research is the total index of the Tehran Stock Exchange from 1991 to 2021. In this research, the tools used to measure and measure the desired variables are the documents and statistics of the Tehran Stock Exchange and for data analysis. This research uses descriptive statistics and inferential statistics of Ks, t, r, and (Dickey-Fuller Test) tests, as well as a multi-layer perceptron neural network algorithm. Tehran compared to other linear methods and the fact that the designed neural network model has the power to predict the total index up to 1.7% error, also the Tehran Stock Exchange stock index follows a non-linear process and finally evaluates the results of the model at the end of the suggestions. An application was presented for users and researchers.