PREDICTION OF STOCK MARKET USING LSTM NEURAL NETWORKS

Abstract

This thesis explores the use of neural networks and Long Short-Term Memory (LSTM) techniques for stock prediction. The focus is on understanding the potential of these techniques in forecasting stock prices by analyzing financial data and developing predictive models. The study will evaluate the effectiveness of LSTM models in comparison to traditional time-series models and assess the impact of different hyper parameters on the accuracy of stock predictions. The aim is to provide insights into the advantages and limitations of using LSTM and Neural networks for stock prediction and to provide guidelines for practitioners and researchers in the field.