Stock Market Prediction with LSTM-based Neural Networks: An Empirical Analysis

Vedant  Pawar; Pallavi  Mangrulkar; Niranjan  Taware; Vinayak  Shirke

Abstract

A stock market is a composite of markets and exchanges where the buying and selling of publicly traded company stocks occur on a regular basis. It serves as a marketplace for the stocks of corporations that are available to the public. Companies opt for an initial public offering (IPO) on the primary market as a means to generate capital. People buy stocks primarily in the hope that they may rise in price in the future. However, there is always uncertainty in the stock market, and people are reluctant to invest their money in the stock market. As a result, we require technology that can forecast stock values so that investors can invest in the finest stocks. As a web application, this project deals with stock market price prediction using LSTM models and uses Dash to visualise stock market information including actual and forecast values. LSTM, which stands for Long Short-Term Memory, is an iterative neural network designed specifically to capture and understand long-term dependencies. It is commonly utilized in the processing and forecasting of time series data. Many LSTMs have a chain-like form. Instead of a single neural network layer, we have four interacting layers that communicate with each other in a very specific way. Dash is a great library framework that Python can use to create interactive web application dashboards. To implement dash, we need to install dash components.

Abstract

Keywords