Using Clustering and Factor Analysis in Cross Section Analysis Based on Economic-Environment Factors

Homogeneity of groups in studies those use cross section and multi-level data is important. Most studies in economics especially panel data analysis need some kinds of homogeneity to ensure validity of results. This paper represents the methods known as clustering and homogenization of groups in cross section studies based on enviro-economics components. For this, a sample of 92 countries which produce the most greenhouse gases including CO2, clustered based on 18 criteria. Those criteria reduced to five primary components using factor analysis. Clustering of countries done by HCPC (Hierarchical Clustering on Principal Component) method. All 92 countries were clustered in 7 different groups. For each group properties of countries indicates the homogeneity of each cluster. In cross section analysis with many sections, especially analysis based on panel data, clustering, increases assurance of expected homogeneity and validity of result.