<![CDATA[The main objective of this research is to examine the performance of neural network (NN) to predict stock price. This research propose to use daily closing price of TELKOM share for the observation period. Authors propose to use TELKOM share because it is one of the most liquid and blue chip stock in Indonesian Stock Exchange (IDX). This research use purposive sampling. The authors propose to use Artificial Intelligence Neural Network (ANN) to predict stock price. From literature review ANN can predict stock price more accurately than traditional way like linear regression. The Artificial Neural Network (ANN) model use backpropagation algorithm for training session and feed forward NN for testing the output. The authors propose ANN with five nodes outperform ANN with fewer (two) nodes. The reason is that the information contained in five nodes ANN provide more accurate information than two node ANN. Consequently ANN with five nodes can predict stock price better and more accurately than ANN with two nodes. The average percentage error should be less for ANN with five nodes then the percentage error of ANN with two nodes]]>
Copyrights © 2024
