Hybridized Deep Learning Models and Federated Learning Techniques to Forecast Stock Market Movement

View Sample PDF

Author(s): A. Anitha (School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, India), Bidyanand Mishra (Vellore Institute of Technology, India), Rahul Kumar Sahu (Vellore Institute of Technology, India), Nikhil Raj (Vellore Institute of Technology, India), S. Srinath (Vellore Institute of Technology, India)and Balakrishnan Kamaraj (Madurai Medical College, India)
Copyright: 2025
Pages: 38
Source title: Strategic Repositioning in Times of Corporate Crisis: Green Management and Technology Adoption
Source Author(s)/Editor(s): Option Takunda Chiwaridzo (University of Science and Technology Beijing, Beijing, China)and Mufaro Dzingirai (Midlands State University, Zimbabwe)
DOI: 10.4018/979-8-3693-5912-9.ch007

Keywords: Business & Management / Business Science Reference / Management Science / Sustainable Business Practices

Purchase

View Hybridized Deep Learning Models and Federated Learning Techniques to Forecast Stock Market Movement on the publisher's website for pricing and purchasing information.

Abstract

Traditional stock market prediction methods often rely on assumptions and conventional approaches, but the emergence of deep learning techniques offers a transformative opportunity. This study evaluates the effectiveness of several deep learning architectures, including LSTM networks, CNNs, GRU, and Bi-directional GRU and Federated Learning in forecasting stock prices. Through rigorous experimentation and assessment using key metrics such as Loss, MAE, MSE, MAPE, and RMSE, deep learning models, particularly LSTM networks and CNNs, demonstrate notable improvements in accuracy compared to traditional methods. Additionally, the study explores the interpretability of these models, providing insights into the complex dynamics influencing stock price fluctuations. The study aims to predict future stock prices of Cipla by integrating various financial indicators from annual reports to offer investors reliable insights for informed decision-making in navigating the intricacies of the stock market.

The IRMA Community

Research IRM

Hybridized Deep Learning Models and Federated Learning Techniques to Forecast Stock Market Movement

Purchase

Abstract

Related Content

IRMA Sponsors