Benchmarking Türkiye’s Sustainable Development Performance: SDG Interactions, Macroeconomic Predictors, and Explainable Machine Learning Evidence from Selected Major Muslim Economies
Sustainability (Switzerland), cilt.18, sa.15, 2026 (SCI-Expanded, SSCI, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 18 Sayı: 15
- Basım Tarihi: 2026
- Doi Numarası: 10.3390/su18157716
- Dergi Adı: Sustainability (Switzerland)
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Social Sciences Citation Index (SSCI), Scopus, CAB Abstracts, Geobase, INSPEC
- Anahtar Kelimeler: economic development, explainable machine learning, major Muslim economies, SHAP, sustainable development goals (SDGs), Türkiye
- Yıldız Teknik Üniversitesi Adresli: Evet
Özet
Sustainable development has become a strategic priority for countries seeking long-term economic prosperity while balancing social, environmental, and institutional objectives. According to the April 2026 International Monetary Fund World Economic Outlook projections, Türkiye is expected to become the largest economy among the selected major Muslim economies in terms of nominal gross domestic product. Motivated by this development, this study benchmarks Türkiye’s sustainable development performance against nine comparable economies and examines the relationships between selected Sustainable Development Goal (SDG) indicators and economic performance. The analytical framework integrates benchmarking, correlation, SDG network analysis, regression modeling, and explainable machine learning using SDG and macroeconomic data. The findings reveal both synergies and trade-offs among the selected SDGs. Strong positive relationships emerge between SDG12 and SDG13 and between SDG4 and SDG16, confirming the interconnected structure of sustainable development. Türkiye ranks second in overall SDG performance and performs particularly strongly in clean energy, innovation and infrastructure, and international partnerships. Regression results identify education, clean energy, innovation, and climate action as statistically associated with economic performance. Under a strict year-based chronological holdout and rolling-origin validation framework, the tuned Random Forest model achieved the strongest out-of-sample predictive performance. SHapley Additive exPlanations analysis indicates that one-year lagged economic performance is the dominant predictor, while current macroeconomic conditions and selected SDG indicators provide additional predictive information. The findings demonstrate substantial temporal persistence in economic performance and show that sustainability indicators complement prior economic information in forecasting. The study contributes to sustainability research by integrating comparative assessment, systems-oriented analysis, econometric modeling, and explainable machine learning while clearly distinguishing statistical association from predictive contribution.