مدل سازي پيش بيني قيمت سهام با استفاده از شبكه عصبي و مقايسه آن با روشهاي پيش بيني رياضي

Modelling of Prediction Stock Price by Using Neural Networks and Compare it with Mathematical Prediction Methods

Different methods for prediction of future situation always have been one of the important concerns of scholars of different science .Naturally in this way; those methods would be more sustainable and applicable which have minimum errors in prediction. During last years, many mathematical methods like: simple average, weighted average, double average and regression have been accepted and applied, but in some situations it had some problems as welt. With the application of artificial intelligence like neural networks, specifically when there is proper relation between data, dependent and independent variables, a lot of hope raised and in a way that the replacement of neural networks models with mathematical methods. In this paper, the application of neural networks models and regressions in the prediction of stock prices have been evaluated and the prediction errors of these models have been measured. The research method which has been used in this paper, is evaluation method.