The relevance of the research topic is determined by the importance and role of the Arctic potential for solving Russia's strategic tasks in modern conditions of challenges and threats, as well as the possibilities of modeling in forming a qualitative basis for managerial decision-making to improve the effectiveness of public administration. Fundamentally new infrastructure and production solutions are being implemented in the Arctic macro-region, which can be scaled in the future, which determines the importance of modeling the development of the Arctic zone of the Russian Federation based on modern data analysis methods. When modeling the development of the Arctic macro-region, it is necessary to take into account such features as the limited and fragmented information collected, as well as the complexity of integrating heterogeneous data (economic, social, environmental, etc.). In this regard, the implementation of a set of modeling tasks based on modern data analysis methods requires various approaches (econometric modeling, cognitive technologies, machine learning, and big data analysis methods) that allow analyzing complex socio-economic, environmental, and infrastructural processes. The combination of various methodological approaches makes it possible to ensure the accuracy of the model, which can be used in developing strategies for the sustainable development of Arctic territories, planning infrastructure projects and making management decisions. The aim of the study is to explore the possibilities of modeling the development of the Arctic macro-region using modern data analysis methods. The aim defined the objectives of the study: to analyze the results of research in this subject area; to consider the clustering method (cluster analysis) as one of the effective methods of substantiating management decisions on the implementation of the Development Strategy of the Arctic zone of the Russian Federation; to identify promising areas of future research. The work used a systematic approach, logical analysis, synthesis, open source content analysis, regression analysis, and cluster analysis. The information base was compiled by Rosstat data on the Arctic regions for the period 2015–2023. As a result of the study, the expediency of using the hierarchical clustering procedure implemented using the JASP data analysis program is substantiated. During the cluster analysis, all the Arctic regions of Russia were grouped into two clusters based on the proximity of specific GRP values, which allows for subsequent regression analysis within each cluster to obtain more accurate results. As a promising area of research, the use of synthetic control methodology is proposed, which makes it possible to create an alternative scenario for the development of a macro-region for comparison with real development and assessment of the economic effect of implementing a set of strategic decisions of the state. The scientific novelty of the study is to improve the approach to modeling the development of the Arctic macro-region using predictive (predictive) analytics methods such as regression analysis, time series method, clustering. The practical significance of the results is determined by the possibility of their application by public authorities and management to develop forecasts for the development of the Arctic zone
Keywords
regional development, Arctic macro-region, modeling, forecast, econometric model, data, predictive analytics, cluster analysis