Modeling economic security of Russian regions using correlation, PCA and clustering methods

In the face of growing economic challenges, assessing the resilience of Russian regions is becoming increasingly important. The aim of this study is to model the level of economic security based on a formalized analysis of key socio-economic indicators. Methods used include correlation analysis, normalization, principal component analysis (PCA), and KMeans clustering. As a result, a typology of eight regions by resilience level was developed, revealing correlations among poverty, unemployment, income, and investment. The study has a practical focus and can support analytical tools for strategic planning and regional risk assessment. Aim. The main goal of the study is to develop a formalized model for assessing the level of economic security of Russian regions. This involves the structural analysis of interconnected socio-economic indicators reflecting regional development and the subsequent classification of territories according to their stability and vulnerability levels. Methods. The methodology combines several analytical techniques for multidimensional data processing. Pearson correlation analysis is used to explore interdependencies among variables, followed by normalization procedures and principal component analysis (PCA) to reduce data dimensionality while preserving key information. Finally, KMeans clustering is applied to classify regions into homogeneous groups based on structural similarities. Results. Based on official statistical data for 2022, a classification of eight regions of the Russian Federation was carried out according to the level of economic stability, and stable interdependencies between socio-economic indicators were identified. The study included the selection and justification of indicators that reflect the state of regional resilience, the construction of a correlation matrix to explore relationships between variables, dimensionality reduction using principal component analysis (PCA), and clustering of the Russian regions using the KMeans algorithm to form a typology based on economic security levels. The results were interpreted with regard to the structure of the data, enabling conclusions about the resilience and development profiles of the analyzed regions. Conclusions. The results of the study are of high practical significance and can be applied in the development of differentiated regional socio-economic policies, as well as in strategic planning and decision-making under macroeconomic uncertainty. The constructed clustering model accounts for structural differences across regions, while the identified relationships between indicators contribute to building more accurate forecasts of regional resilience. The methodological approach used in this research can be scaled to larger groups of regions and adapted for different time frames to monitor changes in economic sustainability over time.