KANNAMMAL, V and Gayathri, B (2025) Machine Learning and Big Data Analytics in Stock Market Prediction: An Empirical Study of NSE. Machine Learning and Big Data Analytics in Stock Market Prediction: An Empirical Study of NSE, 5. pp. 1575-1584. ISSN 2198-4182

[thumbnail of VK- Spinger Publication- Big Data Analytics and Stock Market Performance.pdf] Text
VK- Spinger Publication- Big Data Analytics and Stock Market Performance.pdf - Published Version
Download (1MB)

Abstract

The exponential growth in digital financial data has fundamentally transformed the structure and functioning of modern stock markets. The integration of Big Data Analytics (BDA) with Artificial Intelligence (AI) and Machine Learning (ML) has enabled market participants to process massive volumes of real-time financial information to forecast prices, manage risks, and optimize portfolio decisions with unprecedented accuracy. The Indian stock market, particularly the National Stock Exchange (NSE), witnessed historic expansion during the 2024–2025 financial year, with market capitalization crossing ₹438.9 lakh crore and over ₹18.7 lakh crore mobilized through equity and debt instruments.
The present study examines the effectiveness of Big Data Analytics in predicting stock market performance using NSE data for the financial year 2024–2025. The study adopts a descriptive and analytical research design using both primary data from investors and secondary data from NSE indices, trading volumes, corporate disclosures, macroeconomic indicators, and financial news sentiment. Machine learning-based predictive models are compared with traditional time-series models to evaluate forecasting accuracy, risk management efficiency, and investment decision quality.
The findings reveal that Big Data-driven models significantly outperform traditional methods in terms of predictive accuracy, volatility handling, and risk-adjusted return optimization. The study also highlights major implementation challenges such as data noise, overfitting, cybersecurity risks, lack of model explainability, and regulatory concerns. The paper concludes with practical recommendations for policymakers, stock exchanges, financial institutions, and investors to promote ethical, secure, and data-driven capital market ecosystems

Item Type: Article
Uncontrolled Keywords: Big Data Analytics, NSE, Stock Market Forecasting, Machine Learning, Artificial Intelligence, Financial Technology, Market Volatility, Investment Decision-Making
Divisions: PSG College of Arts and Science > Department of Commerce
Depositing User: Dr. B Sivakumar
Date Deposited: 19 May 2026 04:01
Last Modified: 19 May 2026 04:01
URI: https://ir.psgcas.ac.in/id/eprint/2842

Actions (login required)

View Item
View Item