Skip to content
Conference Open access

Analyzing gold price fluctuations based on stock market movements in NSE using machine learning models of long short-term memory (LSTM), convolutional neural networks (CNNs), and gated recurrent units (GRUs)

2026 · EPJ Web of Conferences · 0 citations · 13 references

Abstract

This research analyzes gold price oscillations on the basis of stock market movements in the National Stock Exchange (NSE) by applying machine learning models of Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNNs), and Gated Recurrent Units (GRUs). Using historical data from 2015-2025, the findings show that while LSTM demonstrates strong forecasting capability, the GRU model achieves superior accuracy across major error metrics, including an MAE of 1.23, MSE of 2.45, RMSE of 1.57, MAPE of 1.12%, and sMAPE of 1.48%. CNNs perform well in predicting short term directional trends, making them suitable for trading-oriented strategies. The results highlight the influence of stock market performance and key economic indicators on gold prices and confirm the effectiveness of deep learning methods in financial forecasting. The study further recommends the development of hybrid deep learning frameworks, incorporation of real-time macroeconomic variables, and advanced feature engineering techniques to enhance model robustness and forecasting reliability. These insights offer valuable implications for investors, financial analysts, and policymakers seeking more intelligent and data-driven tools for financial market prediction and strategic decision-making.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.