Economics Research

Forecasting exchange rates of selected currencies (Euro to Dollar,, Dollar to Yuan, Bitcoin) and gold using neural networks and metaheuristic algorithms (Genetic and Grey Wolf Optimization)

Articles in Press

Document Type : Research Paper

Authors

Azar Azari gharelar 1
Mahmood Khoddam 2
Maryam Gavara 3
Mohammad Maghasedi 4

1 bank mellat-PhD Student in Financial Engineering

2 Assistant Professor, Department of Financial Management, Islamic Azad University, Karaj, Iran

3 Assistant Professor, Department of Accounting, Islamic Azad University, Karaj, Iran

4 Associate Professor Department of Mathematics, Islamic Azad University, Karaj, Iran

https://doi.org/10.22054/joer.2026.91266.1330
Abstract
The aim of this study was to predict the exchange rates of the Euro to Dollar, Dollar to Yuan, Bitcoin, and gold using neural networks and combining them with metaheuristic algorithms, specifically Genetic Algorithm and Grey Wolf Optimization. This research is quantitative in nature, with data on the mentioned exchange rates, gold, and macroeconomic indicators of the United States from 2010 to 2024 being analyzed on a monthly basis. The relationship between variables is examined to address the research questions. The collected data were analyzed using Excel, EViews, MATLAB and Python software. In addition, this study evaluates the performance of metaheuristic algorithms in predicting the selected exchange rates (Euro to Dollar, Dollar to Yuan, Bitcoin) and gold. The evaluation is based on error metrics such as Mean Squared Error, Root Mean Squared Error, R-Squared, and Mean Absolute Error. The results showed that the combination of neural networks with the Genetic Algorithm outperformed the Grey Wolf Optimization algorithm in most metrics. This combination was able to improve the model’s accuracy and efficiency in predicting target variables by reducing errors and increasing the R-squared value. Overall, the combination of neural networks with the Genetic Algorithm provided the best balance between prediction accuracy and model stability when compared to using neural networks individually or combining them with the Grey Wolf Optimization algorithm.

Keywords

Exchange rates
neural networks
metaheuristic algorithms
Genetic Algorithm
Gray wolf Algorithm
Subjects
Economics Research

Articles in Press, Accepted Manuscript
Available Online from 20 September 2026

  • Receive Date 31 March 2026
  • Revise Date 29 July 2026
  • Accept Date 19 September 2026

Home

Editorial Board

Guide for Authors

Submit Manuscript

Contact Us