Methodology of application of methods of multidimensional and dynamic analysis when studying living standards of the population

Abstract

The article presents fundamental approaches to the study of periods of development of socio-economic indicators and their mutual influence. The forms of influence of indicators on each other are investigated. The dynamic analysis of the standard of living of the population and the factors of social and economic spheres is completed with the tools of econometric modeling and canonical analysis. Birth rate, mortality, employment, unemployment, investments in fixed capital, GRP per capita, the account of resource production, the standard of living of the population and fixed assets according to the data of the Samara Region for the period 20062019, registered in the annual collections of state statistics bodies, are considered as indicators. The predicted values of the standard of living by various methods are calculated, confidence intervals for the studied indicators are constructed. By means of adaptive forecasting using the Brown model, forecast values are calculated and confidence intervals are constructed. Using the tools of canonical analysis, integral indicators are calculated and grouping by time factor is carried out. The spatial grouping of the time factor depending on the standard of living of the population and canonical integral factors is presented. According to the results of the analysis of autoregressive models, it was found that in terms of employment, unemployment, fertility, mortality, investment in fixed assets, GRP per capita, resource production and fixed assets, the impact of the indicator of the previous year is statistically significant, and in terms of the standard of living of the population statistically insignificant. In the second-order autoregression, it was found that all statistical indicators have an impact on the studied indicator, except for indicators of employment and the standard of living of the population. Thus, the forms of models of multiple linear regression, paired linear regression and autoregressive models allow us to assess the numerical impact of all indicators on the studied indicators, as well as their impact on the Standard of living of the population. Visualization of multidimensional data contributes to an in-depth analysis of indicators when grouping, for example, by the time factor.