Stock Market Trend Prediction Using Neural Networks and Fuzzy Logic

Stock Market prediction is one of the hottest fields of research due to its commercial applications and the attractive benefits it offers. Nevertheless, the major challenge confronting stock investors is forecasting price movements in stock markets. For these reasons, this thesis presents a stock trend prediction model, generating a signal of either up, constant or down for each trading day based on the historical price of the stock only. The work goal is to give as many accurate predictions as possible. The presented model is a development of an integrated stock market trend prediction model based on Artificial Neural Network and fuzzy logic rules. This study constructs a hybrid model utilizing technical analysis tools: technical indicators and Elliott’s wave theory. The presented model will be conducted by developing predictor system, consisting of three main phases. In the first phase a trend predictor model based on neural network is developed. The Second phase is a fuzzy rule based system to predict the short term stock trend. The third and last phase is integration between the first two phases using neural network. This phase work on generating the final output.