Deep Learning Tools for Predicting Stock Market Movements

1Design and Development of an Ensemble Model for Stock Market Prediction Using LSTM, ARIMA, and Sentiment Analysis

Poorna Shankar^1*, Kota Naga Rohith² and Muthukumarasamy Karthikeyan³

¹Department of MCA, Indira College of Engineering and Management, Pune, India

²Salesforce Consultant, Essen, Germany

³Chemical Engineering and Process Development, National Chemical Laboratory, Pune, India

Abstract

The accurate prediction of stock market movements is a challenging task due to its complex and dynamic nature. In recent years, machine learning techniques have shown promise in addressing this challenge. This study focuses on the design and development of an ensemble model that combines long short-term memory (LSTM), autoregressive integrated moving average (ARIMA), and sentiment analysis to enhance stock market predictions. The ensemble model leverages the strengths of LSTM, which captures long-term dependencies in sequential data, and ARIMA, a statistical model known for its ability to capture linear and autoregressive relationships in time series data. Additionally, sentiment analysis is incorporated to analyze and quantify the impact of public sentiment expressed in textual data on stock market dynamics. The research methodology involves collecting historical stock market data, sentiment analysis data, and performing preprocessing steps to ensure data quality. The LSTM and ARIMA models are developed and trained using the collected data. Sentiment analysis techniques are applied ...