UDC 004
DOI: 10.36871/2618-9976.2022.05.008

Authors

Zhukov R.A.
Candidate of Physical and Mathematical Sciences, Associate Professor, Associate Professor of the Department of Mathematics and Computer Science, Financial University under the Government of the Russian Federation (Tula Branch), Tula, Russia
Prokopchina S.V.
Doctor of Technical Sciences, Professor, Financial University under the Government of the Russian Federation, Moscow, Russia
Giniatov I.A.
4rd Year Student of the Direction "Automation of Technological Processes and Productions", Bauman Moscow State Technical University, Moscow, Russia
Nikolina E.M.
4rd Year Student of the Direction "Automation of Technological Processes and Productions", Bauman Moscow State Technical University, Moscow, Russia

Abstract

The article presents a module for generating large samples with a given distribution law based on small samples with unknown distribution types using Bayesian mathematical statistics. For samples obtained from an external medium or from another program, the coefficients of asymmetry and kurtosis projected onto the plane of Pearson moments are considered. Classes of distributions with corresponding dividing boundaries are defined on the plane of moments. When a point with coordinates equal to the values of the coefficients of asymmetry and kurtosis is hit, the most probable distribution class is determined, as well as the nearest – least probable classes. For the selected class, a large sample is generated by the volume defined by the user. The main characteristics of the distribution are determined by the sample. Graphs are plotted for the most and least probable distributions, and the data obtained (the sample, its main characteristics) are output to a file that is transmitted for further processing to an external program. The software module can be used both when working with technical systems and when modeling socioeconomic processes in order to obtain correct results.

Keywords

Bayesian mathematical statistics, Pearson plane, Probability distribution density, Distribution class, Statistical sampling, Socio-economic systems, Technical systems, Software module