Financial Performance Evaluating and Ranking Approach for Banks in Bist Sustainability Index Using Topsis and K-Means Clustering Method

Abstract

The purpose of this study is to identify, analyze, and evaluate the differences between banks in the Borsa İstanbul (BIST) Sustainability Index and other banks based on their financial performance. This study employs an approach based on a technique for order of preference by similarity to an ideal solution to evaluate and rank 46 Turkish banks with an evaluation framework consisting of 11 financial ratios between 2015 and 2019. The entropy, equal weight, standard, and variance methods were adapted to determine the weights of the financial ratios. Ultimately, the closeness coefficients of the banks derived from the four different weighted TOPSIS methods assist in identifying the positions of banks within other banks using K-means clustering analysis. However, this analysis has two main drawbacks. One is to determine the value of k and the other is to select the initial centers. The Calinski-Harabsz Index (CHI) was used to determine the validity of k. To solve the initial center drawback, we ran the clustering algorithm for all combinations of initial centers from the dataset. CHI is again used to determine which cluster group, derived from a different initial center, is more accurate. Finally, we present the results obtained by using this process for a set of 46 banks.