Comparison of various static and dynamic artificial neural networks models in predicting stock prices

Subject Areas : Financial Knowledge of Securities Analysis

علی اکبر نیکواقبال ¹ (ندارد)
نادیا گندلی علیخانی ² (ندارد)
اسماعیل نادری ³ (مسئول مکاتبات)

Received: 2013-08-24 Accepted : 2013-10-12 Published : 2014-06-15

Keywords: Keyword: Prediction, Stock Market, ANFIS Model, ANN Model, NNAR Model,

Abstract :

AbstractIn this disquisition, has been paid to comparing the performance of static anddynamics neural network by purpose choosing appropriate model in predicting of TehranStock Exchange. The data used in this study consists of daily and interval of time1388/1/5 to 1390/8/30, that Including 616 observation for in sample and out of sampleforecasting. Approximately 90% of these observations (556 data) use to estimatecoefficients of the model and the rest of them (60 data) use to forecast out of sample.Models are also employed in this research; two stationary neural network models such asfuzzy neural network (ANFIS) and artificial neural network (ANN) and a dynamicregression neural network model (NNARX). The results of this survey indicate thatAccording to Criteria to calculate the forecast error, among Mean squared error (MSE)and root mean square error (RMSE), Fuzzy neural network model of static, dynamicregression models, neural networks, and finally static artificial neural network modelshave lowest prediction error, Respectively.