Forecast of Borsa Istanbul Dividend Index (XTMTU) Trend by Using the Method of Artificial Neural Network

Abstract

Use of artificial neural networks (ANNs) in the field of finance contributes to the solution of even the most complex problems by increasing the efficiency and the speed in decision making. Hopeful results have been obtained in the prediction of stock exchanges from the studies conducted by using artificial neural networks (ANNs) just like the other financial forecasts. In this study, we forecast the daily closing values between 03.01.2017 and 07.05.2017 by using the daily closing values of the Borsa Istanbul Dividend Index (XTMTU) betwen 02.03.2014 and 09.03.2017. Our main object in this research is to test the predictabiliy power of artificial neural networks (ANNs) by using the results obtained on the Borsa Istanbul Dividend Index (XTMTU). The results of our study indicate that the artificial neural networks determine the trend of the Borsa Istanbul Dividend Index (XTMTU) far better than many statistical and traditional methods.

Keywords: Artificial Neural Networks (ANNs), Multilayered Feedforward Network, Borsa Istanbul Dividend Index (XTMTU), financial analysis.