Classification of performance of the Turkish banking sector based on financial indicators with clustering analysis

The most important actor that constitutes the institutional structure of the financial market is the banking sector. Due to its sensitive structure, the banking sector is exposed to sectoral fluctuations and becomes the focus of intense discussions. This study aims to classify commercial banks operating in Turkey with clustering analysis, a multivariate statistical method, in terms of similarities and differences in the risk and financial performance of banks using financial ratios obtained from the balance sheets and income statements. The classification in question was carried out using the Ward method, which is one of the hierarchical clustering analysis methods, the K-means method, which is one of the non-hierarchical clustering analysis methods, and the two-step clustering method, which is a hybrid method as a combination of hierarchical-non-hierarchical methods. As a result of the study, public banks, private banks and banks with foreign capital, without distinction, banks came together and formed very different clusters than their current situation. But the common point of the analyses according to the 3 methods discussed is that public banks are never separated from each other, but there are changes between clusters in private and foreign-owned banks.