АНАЛІЗ ВИТРАТ ДЕРЖАВНОГО БЮДЖЕТУ ПЕРІОДУ ВІЙНИ З ВИКОРИСТАННЯМ ІНСТРУМЕНТАРІЮ ЛОГІСТИЧНОЇ РЕГРЕСІЇ

This article addresses the scientific problem of identifying the capabilities of the logistic regression toolkit in the analysis of government spending, in particular, military spending. This problem is relevant not only for the Ukrainian economy in the active phase of the war, but also for other European countries. Logistic regression is identified as an analytical tool whose potential grows in conditions of economic uncertainty.

In order to test the scientific hypothesis that the logistic regression toolkit creates additional opportunities for economic analysis of spending, a logistic regression model was created based on actual data for 35 years of modern Ukrainian history. The created model makes it possible to answer the question: which variables (predictors) determined the high values of the shares of military spending in Ukraine's GDP. It was found that seven variables were influential, of which three had a predominant impact on high military spending and turned out to be the most statistically significant. The list of the most influential variables includes such as foreign economic assistance to the country, GDP per capita, taking into account purchasing power parity (PPP), and GNI per capita, taking into account PPP.

The study used the methodology of theoretical analysis of government spending, as well as methods of statistical analysis of data series, and the actual tools for building a logistic regression model. The calculations were made on the basis of statistical data on the Ukrainian economy, access to which is provided by the World Bank.

The logistic regression model constructed and presented in this article, which explains the level of military spending in the Ukrainian economy, is qualitative. The quality of the model was checked in several ways. The check showed that the model can create a certain basis for making management decisions on ensuring high military spending in the Ukrainian economy. It can also be taken into account when forecasting changes in military spending, under different scenarios of changes in influential factors.