UDC 004.94, 519.862
DOI: 10.36871/2618-9976.2022.11.006
Authors
Sergey A. Zadadaev
Candidate of Physical and Mathematical Sciences, Associate Professor, Head of the Department
of Mathematics, Financial University under the Government of the Russian Federation, Moscow,
Russian Federation
Pavel B. Lukyanov
Doctor of Economical Sciences, Professor, the Department of Mathematics, Financial University under
the Government of the Russian Federation, Moscow, Russian Federation
Abstract
The aim of the work was to model the mechanism of the influence
of an exogenous resource variable on the predicted indicator
of the national development goal, in which one model was
selected from an infinite family of hypothetical resource allocation
models within a year, reflecting the uncertainty inherent in
soft computing.
The study of problems based on small sample data in the age of
big data is substantiated by the fact that there are a number of
longitudinal tasks of longterm
research in which, after 5–6
years of testing any methods, it is necessary to obtain data on
their effectiveness and predictability in general. At the same
time, the data slice occurs strictly at the end of the year as a control
and summary measurement. In such cases, special econometric
specifications of models are required that take into account
the "physics" of relationships. This situation is similar
to the situation in mechanics, when small data are compensated
by the laws of motion and play the role of initial data in the Cauchy
problem for differential equations – when the law itself
is known and does not need to be estimated using big data.
Keywords
Small data, Forecast indicators of national development goals, State program "Development of education", National project "Education", Alternative forecasting indicators, Resource allocation models
