Research Article

CLASSIFICATION OF PROVINCES USING CLUSTERING METHODS IN TERMS OF BANK DEPOSITS AND LOANS

Volume: 27 Number: 2 April 26, 2026
TR EN

CLASSIFICATION OF PROVINCES USING CLUSTERING METHODS IN TERMS OF BANK DEPOSITS AND LOANS

Abstract

Numerous studies in the literature classify provinces in Turkey using various indicators such as socio-economic development, education, and health, and examine clustering structures based on similarities or differences. This study differs by classifying provinces according to bank deposits and bank loans and evaluating their clustering patterns. Six variables related to deposits and loans were used, and the data were obtained from the 2022 statistics of the Turkish Banks Association. Per capita bank deposits and loans were calculated using 2022 provincial population figures, and clustering analyses were conducted based on these per capita values. As the number of clusters, six tiers defined in the SEGE2017 socio-economic development rankings of the Ministry of Industry and Technology were adopted. Additionally, changes in clustering structures were examined for different cluster counts. Although various classification studies exist, this study represents a new attempt in terms of socio-economic development levels, as it evaluates provinces according to SEGE2017 while using financial indicators. For clustering, k-means, complete linkage, Ward’s method, and median clustering were applied. Across all methods, results based on both per capita deposits and loans exhibit notable similarities regardless of the number of clusters. Istanbul and Ankara consistently appear together in a single cluster, while Izmir, Antalya, Bursa, and Mugla are grouped in another. This may be due to the concentration of population in these provinces and their higher level of development compared to other provinces. Most provinces fall into two broad categories and are therefore not compatible with the SEGE2017 provincial classification structure.

Keywords

References

  1. Açıkalın; S. (2023). Covid-19 Pandemi Döneminde Türk Bankacılık Sektöründe Yer Alan Bankaların Kümeleme Analizi İle Sınıflandırılması, (Yayımlamamış Yüksek Lisans Tezi), Balıkesir Üniversitesi, Balıkesir.
  2. Aggarwal, C.C. & Reddy, C.K. (2014). Data Clustering Algorithms and Applications. CRC Press Taylor&Francis Group, New York.
  3. Akgöz, E. (2010). Türkiye’de Ticari Bankaların Finansal Oranlar Yardımıyla Sınıflandırılması: Kümeleme Analizi Yaklaşımı, (Yayımlanmamış Yüksek Lisans Tezi), Kahramanmaraş Sütçü İmam Üniversitesi, Kahramanmaraş.
  4. Albayrak, A. S. & Savaş, F. (2015). Türkiye’de İllerin Sosyo-Ekonomik Gelişmişliğinin Belirleyicileri ve 2012 Yılı Sosyo-Ekonomik Gelişmişlik Sıralaması. AİBÜ Sosyal Bilimler Enstitüsü Dergisi, 15(3), 1-40.
  5. Alpar, R. (2020). Uygulamalı Çok Değişkenli İstatistiksel Yöntemler (6th Ed.). Detay Yayıncılık, Ankara.
  6. Anderberg, M. R. (1973). Cluster Analysis For Applications, Academic Press, New York.
  7. Aydın, D. & Başkır, M. B. (2013). Bankaların 2012 Yılı Sermaye Yeterlilik Rasyolarına Göre Kümeleme Analizi ve Çok Boyutlu Ölçekleme Sonucu Sınıflandırılma Yapıları. Bankacılık ve Sigortacılık Araştırmaları Dergisi, 1(5-6), 29-47. https://doi.org/10.1501/bsad_0000000015
  8. Ayrıçay, Y. & Akgöz, E. (2014). Ticari Bankaların Finansal Oranlar Yardımıyla Sınıflandırılması: Kümeleme Analizi Yaklaşımı. Journal Of Social and Humanities Sciences Research, 1(1), 1-23. https://doi.org/10.26450/jshsr.5

Details

Primary Language

English

Subjects

Econometric and Statistical Methods

Journal Section

Research Article

Publication Date

April 26, 2026

Submission Date

February 10, 2026

Acceptance Date

April 6, 2026

Published in Issue

Year 2026 Volume: 27 Number: 2

APA
Öner, B. (2026). CLASSIFICATION OF PROVINCES USING CLUSTERING METHODS IN TERMS OF BANK DEPOSITS AND LOANS. Cumhuriyet Üniversitesi İktisadi Ve İdari Bilimler Dergisi, 27(2), 461-489. https://doi.org/10.37880/cumuiibf.1885630

Cumhuriyet University Journal of Economics and Administrative Sciences is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY NC).