مجلة المنهل الاقتصادي
Volume 9, Numéro 1, Pages 775-788
2026-05-25

Gold Price Forecasting Using Stacked Lstm Neural Networks: A Deep Learning Approach



Authors : Belgat Amira .

Abstract

Gold plays a unique role in financial markets as a reserve of value, hedge against inflation, and safe haven. Accurate forecasting remains challenging due to complex, non-linear dynamics. This study investigates stacked LSTM networks for gold price prediction using daily data from January 2023 to December 2025. The model incorporates seven variables and three stacked LSTM layers with dropout regularization. Performance is evaluated using Mean Squared Error, Mean Absolute Error, and R-squared, achieving training R² of 0.9665 and testing R² of 0.5646. Residual analysis confirms unbiased predictions with approximately normal errors. A 30-day forecast with confidence intervals quantifies uncertainty. Findings show stacked LSTM effectively captures short- and long-term dependencies, offering actionable insights for traders, investors, and risk managers.

Keywords

LSTM ; gold price prediction ; deep learning ; time series forecasting ; financial markets

Les articles similaires

Predictive Modeling Of Ozone Dosing In Drinking Water Treatment Plant Using Deep Learning Comparative Study Between Deep Neural Networks And Convolutional Neural Networks

Hellal Aouatef .  Djeddou Messaoud .  Loukam Imed .  A. Hameed Ibrahim .  Al Dallal Jehad .  Shawaqfah Moayyad . 
pages 69-83.

Gold Price Prediction Based On Arima And Lstm Neural Network

Mohammed Rouaba . 
pages 134-150.

Forecasting Government Final Consumption Expenditure In Algeria: Comparative Performance Of The Arima (box-jenkins) Model And Artificial Neural Networks (mlp & Lstm) Up To 2034

Lagoune Salim . 
pages 13-27.

Intrusion Detection Based On Neural Networks Neural Networks In Iot

Chaima Bensaid . 
pages 29-33.

Objective Evaluation Of The Pathological Voice Based On Deep Learning Neural Networks In An Algerian Hospital Environment

Kabache Mahraz .  Guerti Mhania . 
pages 47-51.

Formats de citation