An examination on classification of OECD countries by clustering and discriminant analysis

Cluster analysis is a multivariate statistical analysis method that assigns similar data to clusters, that is, assigns units close to each other when similarity is high within groups, and assigns distant units to clusters between groups. Discriminant analysis is one of the multivariate statistical analysis techniques, the main purpose of which is separation, determining the groups to be assigned by benefiting from the characteristics of the observations, or finding the function that will separate the previously determined groups from each other. In this study, 37 OECD countries, based on 2020 with discriminant and cluster analysis, economic growth, military expenditures, education expenditures, health expenditures, subsidy and transfer expenditures, and investment expenditures are classified in terms of variables. In the cluster analysis, Belgium, Netherlands, Austria, Finland, France, England, Canada, Germany, Switzerland, Norway, Sweden, Australia, New Zealand, Spain, Japan and the USA formed a cluster; Hungary, Slovenia, Italy, Portugal, Greece, Colombia, Israel, South Korea, Chile, Denmark, Iceland, Mexico, Ireland and Turkey formed a second cluster. As a result of the discriminant analysis, it was seen that all 16 countries in the first cluster were classified correctly and all 21 countries in the second cluster were correctly classified. In terms of public expenditures, transfer subsidy expenditures and health expenditures that are higher than in other countries are clustered in the same group. It has been observed that there is a positive relationship between the rate of economic growth that should be under normal conditions and the rate of public expenditure.