Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

---

Financial sanctions, by restricting access to funds and disrupting trade, have exerted profound effects on Iran’s macroeconomic performance, largely due to the economy’s dependence on oil exports. This study, for the first time in Iranian economic literature, applies advanced machine learning models—Logistic Regression, Random Forest, XGBoost, and LightGBM—combined with SHAP analysis to evaluate the effectiveness of monetary and fiscal policies in mitigating these adverse effects. Using monthly data from 2006 to 2024 and a time-based validation approach, the models were trained to forecast policy success. Results show that XGBoost achieved the highest predictive performance with 91% accuracy and an ROC-AUC of 0.94, followed by LightGBM (88%, 0.92), Random Forest (85%), and Logistic Regression (75%). SHAP interpretation identified the real interest rate, tax-to-GDP ratio, government spending, liquidity growth, and sanctions dummy as the most influential factors shaping policy outcomes. The findings confirm that tree-based algorithms outperform traditional models in capturing complex nonlinear relationships in economic policy prediction. Overall, this research highlights the potential of data-driven approaches to guide the formulation of targeted, evidence-based policies in sanction-hit economies, offering a robust framework for future macroeconomic policy design under external constraints.

Keywords: Machine Learning, SHAP, Financial Sanctions, Monetary Policy, Fiscal Policy

     

Add your comments about this article : Your username or Email:

---