Paper Title
Stock Market Prediction With Artificial Neural Network- A Systematic Review

Abstract
There have been numerous studies to predict stock market prediction using various ANN techniques. The present study summarizes the various techniques and their effectiveness for stock market prediction. There have been landmark studies such as; (Wei Huanga, 2004) used Support Vector Machine (SVM) to predict the index direction, (Yakup Kara, 2010) compared Artificial Neural Networks (ANN) and SVM in predicting the direction of the Istanbul Stock Exchange, (Mingyue Qiu, 2016) used ANN to forecast the direction of the Japan Stock Exchange. Kimoto et al. proposed a modular ANN to predict the Tokyo Stock Exchange Index (Kimoto, Asakawa, Yoda, & Takeoka, 1990). Mizuno proposed a learning algorithm for ANN to improve the prediction accuracy by neural networks (Hirotaka Mizuno, 1998). In (Pakdaman Naeini, 2010), a predictive model based on Multilayer Perceptron was proposed to predict the Tehran Stock Exchange Corporation (TSEC) index with 1.5% error. In (White, 1998), the price of IBM stock was predicted using feedforward learning. In (Jay Desai, 2013), the closing value of the Indian S&P CNX Nifty 50 Index was predicted using an ANN model. In (C. D. Tilakaratne, 2007) (K. Schierholt, 1996), feed-forward learning and multilayer ANN were used to predict S&P500. In (P. Sutheebanjard, 2010), authors used back-propagation learning in ANN to predict the Thailand Stock Exchange Index. In (JINGTAO YAO, 1999), the back-propagation neural network was used to predict the Kuala Lumpur Composite Index. H. White predicted the IBM daily common stock price using three-layer feed-forward ANN, which consisted of one input layer, one hidden layer, and one output layer (White, 1998). In (Heping Pan, 2005), authors predicted the price of one day ahead based on the current day’s close price using feed-forward and probabilistic ANN. In (Karsten Schierholt, 1996), MLP (Multi-Layer Perceptron) and probabilistic ANN were used to predict S&P 500 index. In (Mehrara, Moeini, Ahrari, & Ghafari, 2010), authors compared back-propagation learning with genetic-based back-propagation learning. In (Feng Li, 2009), authors used a three-layered feed-forward ANN with Levenberg-Marquardt back-propagation algorithm to predict the Shanghai Stock Exchange Index. Quah et.al. compared the performance of MLP feed-forward and back-propagation and Group Method of Data Handling(GMDH) with the Genetic Algorithm in predicting the stock price index in Tehran Stock Exchange (Quah, 2007). The present study presents a systematic review of all above mentioned and other studies and assesses which machine learning and especially ANN techniques have been more effective in stock market prediction. Keywords- Artificial Neural Networks, Stock Market Index, Feed-forward Learning.