Exploring the effectiveness of a multilayer neural network model for gold price prediction

Effective gold price forecasting model is an essential tool for price discovery and benchmarking market changes for mining project across the world. This study presents a model for effective prediction of gold price variation across the world. An experimental analysis was conducted for forecasting monthly US market gold prices from December 1978 to March 2023 using the Autoregressive Integrated Moving Average (ARIMA) model and Multilayer perceptron (MLP) regression model. Optimized hyperparameter search determined the lowest Mean Squared error (MSE) and Mean Absolute (MAE) results with ARIMA (2, 1, 1) and MLP best parameters. The proposed multilayer perceptron (MLP) model demonstrates an improvement in the effective prediction obtained from the experimental analysis.