Intelligent Forecasting of Gold Price Trends in Financial Markets Using LSTM Neural Networks Based on Conservative Trading

Intelligent Forecasting of Gold Price Trends in Financial Markets Using LSTM Neural Networks Based on Conservative Trading

Mahboobeh Mohsen al-Hosseini1 Ali Mokhtarbaf2 Seyyed Javad Mirabedini3

1) M.A. Student, Department of Computer Engineering, Central Tehran Branch, Islamic Azad University, Tehran, Iran |
2) Assistant Professor, Department of Computer Engineering, Central Tehran Branch, Islamic Azad University, Tehran, Iran |
3) Assistant Professor, Department of Computer Engineering, Central Tehran Branch, Islamic Azad University, Tehran, Iran |

IOI Code : AAHD-ABKKEF
Publication : National Conference on Information Technology, Nanotechnology, Artificial Intelligence and Technological Futures Studies(itnaf.ir)
Abstract :
    Predicting the price fluctuations of strategic assets such as gold remains a fundamental challenge in the fields of financial engineering and financial intelligence, due to their high sensitivity to macroeconomic variables, geopolitical events, and complex market mechanisms. This study aims to develop and evaluate a structured deep learning model based on Long Short-Term Memory (LSTM) networks to forecast the directional movement of gold prices (GC=F) and to implement an algorithmic trading system with a conservative risk management approach. To this end, daily gold price data from 2017 to 2026 were extracted and, after preprocessing and feature engineering using the Relative Strength Index (RSI) and rolling standard deviation of returns, were fed into the recurrent neural network. To handle outliers, the Robust Scaler algorithm was employed. Empirical findings from backtesting simulations demonstrate that the AI-driven strategy achieved an approximate 90% return (growing the initial capital from 10,000 to 19,023 units) with a significantly lower Maximum Drawdown (MDD) compared to the traditional buy-and-hold benchmark. Moreover, evaluation of the confusion matrix confirms precise risk control and effective filtering of false positive signals, underscoring the high performance of sequence-based models in highly volatile markets [1].

Keywords : deep learning lstm financial market prediction algorithmic trading risk management