ANN-Based Stock Market Price Prediction Model

Foreseeing stock prices is one of the most challenging and multi-dimensional issues dominated by the characteristics of most of the financial markets, which are constant and unpredictable at the same time. The traditional methods of analysis do not often succeed in explaining the complex nonlinear interdependencies that usually exist in the stock price data. However, with the increase of machine learning acceptance, artificial neural networks have been devised to naturally solve such an issue. This research paper endeavors to formulate a model that builds on artificial neural networks techniques of predicting the future stock prices. The model is fitted on prior data composed of the values of the stock in question and some relevant market indicators in an attempt to explore any existing pattern or trend. The study evaluates the forecasting potential of the ANN model using the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) measures and presents these results against the backdrop of more common forecasting methods. The results of this research aim at increasing the volume of literature related to forecasting with the use of artificial intelligence as well as demonstrating the most realistic boundaries and prospects of this model in constantly changing environments.