Date of Publication :4th April 2024

Abstract:This article provides an exhaustive analysis of stock price prediction models using deep learning based on LSTM neural networks. An advanced sub-type of RNN known as LSTM has risen in importance within financial forecasting due to its ability to detect long-term dependencies and patterns of time series data. The paper discusses different aspects of LSTM-Stock price prediction models including architecture designs, data pre-processing, feature engineering, and evaluation measures. The study evaluates various algorithms including Logistic Regression, SVM, Linear Regression, & LSTM. Additionally, it reveals the issues involving LSTM models when attempting to forecast share price movements. This research adds to the growing literature on utilization of deep learning in predicting stock prices, and can be useful to both scholars and practitioners interested in financial prediction. These outcomes suggest the strength of deep learning methods as powerful means of enhancing decision making in such volatile and uncertained area as stock market.

Reference :