Answers are grounded in this paper and cite their sources.
Forecasting and analyzing the Human Development Index (HDI) at the regional level is an important task in socioeconomic modeling. Although the HDI is calculated based on three indicators, namely life expectancy at birth, level of education, and gross national income per capita, in practice, other factors also affect the final value of the index. The aim of this study is to identify the factors that have the most significant impact on the HDI and to search for linear and nonlinear relationships between the HDI and other indicators. The hypothesis tested is that the poverty rate makes the main contribution to the HDI as a catalyst for many negative phenomena. It is also hypothesized that socioeconomic indicators may have both negative and positive synergistic effects on the HDI. The main research methods are machine learning methods and algorithms for their interpretation. The study is based on data from 255 observations on Russian regions, including 38 socioeconomic indicators. Six machine learning models are used. LIME, Permutation Importance, and SHAP methods are used to interpret the results. The analysis proves that the poverty rate is the most important indicator, exerting a strong negative impact on the HDI in all models. The number of higher education students also shows a consistent positive effect. At the same time, LIME and SHAP methods reveal complex nonlinear relationships, including a negative local effect of the share of high-technology products and gross regional product per capita in the predictions of ensemble models, which indicates the possible existence of threshold and synergistic effects. The identified nonlinearities emphasize the need for a comprehensive approach to managing the development of human potential. The proposed methodology demonstrates high practical value and can be used to analyze the causes of regional differentiation.
No takes yet. Share an insight, caveat, or question.
A 2026 study studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: