Authors : Eman Al-Shamery and Ameer Al-Haq Al-Shamery
Abstract: Stock market prediction is an interest financial topic that has attracted the attention of researchers for the last years. Data mining has been effectively used in financial predicting, hence, researchers have explored technical indicators to optimize the parameters. The main objective of this study is to improve prediction for stock market using a developed method of feed forward neural networks based on optimization model which aims to reduce the error factor depending on the Jacobian vector and Lagrange multiplier according to the converges factor reach to zero. Also, a benchmark of Iraq (Bank of Baghdad) is built and comparted with DOWJONES and S and P500 stock markets. After evaluation stage. The results are compared by K-nearest-neighbors method and decision tree algorithm. The proposed method satisfies better results of prediction according to the accuracy and root mean secured error performance measures.
Eman Al-Shamery and Ameer Al-Haq Al-Shamery, 2018. An Optimized Feed Forward Neural Network for Reducing Error Based Stock Market Prediction. Journal of Engineering and Applied Sciences, 13: 4616-4621.