Comparative analysis of calculation methods factor models of digital economy of Kazakhstan

Abstract

In the scientific article, based on the comparative analysis of the methods for calculating factor
models, a scientific-theoretical generalization and systematization of modern methodological approaches
used in the economic analysis and forecasting of factor models are realized. In particular, economic and
mathematical methods for analyzing time series are considered: simple extrapolation, exponential smoothing
and autoregressive. The calculation of the dependences of statistical indicators on the time factor has been
performed, which makes it possible to assess the level of dynamics of indicators of the digital economy of
Kazakhstan, as well as detailed estimates of the factor indicators characterizing the results of modeling, which
are of interest for analytical calculations. It can be noted that if the forecast by the time series alignment method
is taken as 100%, then the deviations obtained during the study are 7.5%, 5.7%, 5.5%, which in turn indicates a
high level of accuracy of the constructed models.