Identifying Insurance Companies’ Business Models in Ukraine: Cluster Analysis and Machine Learning
a National Bank of Ukraine, Kyiv, Ukraine
b National University of Kyiv-Mohyla Academy, Kyiv, Ukraine

This study examines the performance of the nonlife insurance companies that operated in Ukraine in 2019– 2020. Specifically, we employ a set of clustering techniques, e.g. the classic k-means algorithm and Kohonen self-organizing maps, to investigate the characteristics of the Retail, Corporate, Universal (represented by two clusters), and Reinsurance business models. The clustering is validated with classic indicators and a migration ratio, which ensures the stability of the clusters over time. We analyze the migration of companies between the identified clusters (changes in business model) during the research period and find significant migration between the Reinsurance and Corporate models, and within the Universal model. Analysis of the data on the terminatio of the insurers’ ongoing activity allows us to conclude that companies following the Universal business model appear to be the most financially stable, while their peers grouped into the Reinsurance cluster are likely to be the least stable. The findings of this research will be valuable for insurance supervision and have considerable policy implications.

Full Text
Cite as: Tarnavskyi, O., Kolomiiets, V. (2021). Identifying Insurance Companies’ Business Models in Ukraine: Cluster Analysis and Machine Learning. Visnyk of the National Bank of Ukraine, 252, 37-55.
Citation Format


Abbas, S. A., Aslam, A., Rehman, A. U., Abbasi, W. A., Arif, S., Kazmi, S. Z. H. (2020). K-Means and K-Medoids: Cluster analysis on birth data collected in city Muzaffarabad, Kashmir. IEEE Access, 8, 151847-151855. 

Abolmakarem, S., Abdi, F., Khalili-Damghani, K. (2016). Insurance customer segmentation using clustering approach. International Journal of Knowledge Engineering and Data Mining, 4(1), 

Ahmar, A. S., Napitupulu, D., Rahim, R., Hidayat, R., Sonatha, Y., Azmi, M. (2018). Using K-Means clustering to cluster provinces in Indonesia. Journal of Physics: Conference Series, 1028, 012006. 

Arthur, D., Vassilvitskii, S. (2007). k-means++: the advantages of careful seeding. SODA '07: Proceedings of the eighteenth annual ACM-SIAM symposium on Discrete algorithms (p./pp. 1027--1035), Philadelphia, PA, USA: Society for Industrial and Applied Mathematics.

Bach, M. P., Vlahović, N., Pivar, J. (2020). Fraud prevention in the leasing industry using the Kohonen self-organising maps. Organizacija, 53(2), 128-145.  

Caliński, T., Harabasz, J. (1972). A dendrite method for cluster analysis. Communications in Statistics, 3(1), 1-27. 

Ferstl, R., Seres, D. (2012). Clustering Austrian banks’ business models and peer groups in the European banking sector. Financial Stability Report, 24, 79-95. Austrian Central Bank. Retrieved from 

Honkela, T. (1998). Description of Kohonen’s self-organizing map. In Honkela, T. Self-Organizing Maps in Natural Language Processing. Helsinki: Helsinki University of Technology. Retrieved from 

Kaufman, L., Rousseeuw, P. J. (1990). Partitioning around medoids (Program PAM). In Kaufman, L., Rousseeuw, P. J. Finding Groups in Data: An Introduction to Cluster AnalysisFinding Groups in Data, pp. 68-125. John Wiley & Sons. 

Kohonen, T. (1982). Self-organized formation of topologically correct feature maps. Biological Cybernetics, 43, 59–69. 

Kramarić, T. P., Bach, M. P., Dumičić, K., Žmuk, B., Žaja, M. M. (2017). Exploratory study of insurance companies in selected post-transition countries: Non-hierarchical cluster analysis. Central European Journal of Operations Research, 26(3), 783–807. 

Pedregosa, F., Varoquaux, G., Gramfort, A., et al. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12(85), 2825–2830. Retrieved from 

Rashkovan, V., Pokidin, D. (2016). Ukrainian banks’ business models clustering: Application of Kohonen neural networks. Visnyk of the National Bank of Ukraine, 238, 13-38. 

Rousseew, P. J. (1987). Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. Journal of Computational and Applied Mathematics, 20, 53–65. 

Tibshirani, R., Hastie, T., Narasimhan, B., Chu, G. (2002). Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proceedings of the National Academy of Sciences of the United States of America, 99(10), 6567-6572. 

Velykoivanenko, H., Beschastna, G. (2018). Analysis of the stability and rating of Ukrainian insurance companies. Modelling and Information Systems in Economics, 95, 65-81. Retrieved from 

Vettigli, G. (2019). MiniSom: Minimalistic and Numpy-Based Implementation of the Self Organizing Map (release 2.1.5. 2019). Retrieved from 

Wang X., Keogh E. (2008) A clustering analysis for target group identification by Locality in motor insurance industry. Soft Computing Applications in Business. Studies in Fuzziness and Soft Computing, 230, 113–127. 

Zaqueu, J. R. (2019). Customer Clustering in the Health Insurance Industry by Means of Unsupervised Machine Learning: An Internship Report. University of Lisbon, Information Management School. Retrieved from 

Rights and Permissions
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material.
Submit Your Paper