Insurance Markets’ Research Based on Insurance Indicators Data for 31 Countries and Identification of Russia’s Place in Hierarchical Clustering in R Studio Environment

Abstract

The article is aimed at insurance markets research for 31 countries based on insurance indicators such as insurance density and insurance penetration and providing hierarchical clustering analysis using RStudio. The clustering is done with k-means and Ward methods and the authors tried to interpret the clustering analysis and prove the reasons why countries refer to the specific cluster and explain further development opportunities for each cluster. The authors suggested the hypothesis that Russia should be in one cluster with Eastern European counties and the hypothesis was rejected as a result of the research. The analysis revealed that 31 countries should be divided into 5 clusters and Russia is referred to the cluster with Turkey and Mexico that have very similar insurance systems and share common problems in insurance industry development. The authors infer that insurance instruments that are used by members of one cluster can potentially be adopted by other members within the cluster. The calculations in R are included in the paper and the research can be repeated by students in classes with insurance industry data as well as in other sectors and with different data.