A SHAP-Based Evaluation of Machine Learning Models for CBRT Monetary Policy Decisions
Journal of Economic Policy Researches / İktisat Politikası Araştırmaları Dergisi · 21.08.2026 · 13 (2) · 240-266
DOI: 10.26650/jepr1944519 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Economic Policy Researches / İktisat Politikası Araştırmaları Dergisi, online erschienen am 21.08.2026. Verfasst von Burcu Kartal. Die Arbeit nennt 31 Literaturangaben.
Abstract (Originalsprache)
This study classifies the monthly interest rate decisions of the Central Bank of the Republic of Turkey between 2010 and 2024 into three categories: increase (hike), decrease (cut), and hold, using machine learning methods, and examines which macroeconomic and financial variables are associated with each decision category in a non-causal, prediction-focused manner. Data are obtained from the Federal Reserve Economic Data System and the electronic data delivery system of the Central Bank of the Republic of Turkey. The performance of the XGBoost, LightGBM, and CatBoost algorithms is compared against the Logistic Regression and Naive Bayes baseline models using a walk-forward validation design that preserves temporal order, with all explanatory variables lagged one month relative to the decision date to reduce contemporaneous information leakage. Class imbalance is addressed through class weighting and an adaptive oversampling procedure, and the contribution of each variable is examined using Shapley value-based explanation analysis. None of the three gradient boosting models achieves a Cohen's Kappa value meaningfully above zero in walk-forward out-of-fold evaluation, and the interest rate increase category is not correctly identified by any model; overall performance is comparable to, and in several cases weaker than, a simple majority class baseline. These findings indicate that interest rate decisions cannot be reliably predicted based on the macroeconomic indicators examined, and they support the role of institutional and communication-related factors that are not reflected in these variables. However, these findings alone do not conclusively prove this role.
Zitieren
APA
Kartal, B. (2026). A SHAP-Based Evaluation of Machine Learning Models for CBRT Monetary Policy Decisions. Journal of Economic Policy Researches / İktisat Politikası Araştırmaları Dergisi, 13(2), 240-266. https://doi.org/10.26650/jepr1944519
Vancouver
Kartal B. A SHAP-Based Evaluation of Machine Learning Models for CBRT Monetary Policy Decisions. Journal of Economic Policy Researches / İktisat Politikası Araştırmaları Dergisi. 2026;13(2):240-266. doi:10.26650/jepr1944519
BibTeX
@article{Kartal2026,
author = {Kartal, Burcu},
title = {A SHAP-Based Evaluation of Machine Learning Models for CBRT Monetary Policy Decisions},
journal = {Journal of Economic Policy Researches / İktisat Politikası Araştırmaları Dergisi},
year = {2026},
volume = {13},
number = {2},
pages = {240--266},
doi = {10.26650/jepr1944519}
}
EUVIMED