AL-Qadisiyah Journal For Administrative and Economic sciences

Improving the accuracy of forecasting macroeconomic indicators in Iraq using artificial intelligence compared to traditional models

Document Type : Research Paper

Authors

Sanaa Ali Jabber
Miqdad Jassim Abd
Wadhah R. Rahi
Jehan kassam sahib

University of Al-Muthanna

https://doi.org/10.33916/qjae.2026.01140160
Abstract
The study aimed to demonstrate the extent to which artificial intelligence models can improve the accuracy of forecasting macroeconomic indicators compared to the traditional model used in analyzing these indicators in Iraq for the period 1990-2024. The variables or economic indicators included (GDP, inflation, oil price, exchange rate). The time series was tested for stationarity to apply the ARIMA model compared to artificial intelligence models LSTM and Transformer. Better results were obtained using artificial intelligence models compared to the traditional ARIMA model for all economic indicators. The study recommends the possibility of adopting the artificial model in forecasting instead of traditional models in decision-making and shaping future policies related to the economic aspect of Iraq.

Keywords

forecasting
economic indicators
LSTM
ARIMA
Transformer
Crossmark
AL-Qadisiyah Journal For Administrative and Economic sciences
Volume 28, Issue 1
Spring 2026
Pages 140-160

Home

Guide for Authors

Submit Manuscript

Reviewers

Contact Us