ON THE QUESTION OF DETERMINING THE PROBABILITY DENSITY OF TYPICAL DISTRIBUTIONS FROM EXPERIMENTAL DATA

UDC 519.2; 316.4
DOI: 10.36871/2618-9976.2023.07-2.001

Authors

Svetlana V. Prokopchina,
Doctor of Technical Sciences, Professor, Financial University under the Government of the Russian Federation, Moscow, Russia

Abstract

The article discusses the issues of determining errors in determining the type of distributions. The type of distributions is expressed in the form of probability densities of unimodal distribution laws of the K. Pearson system. Analytical dependences are obtained to determine the necessary volumes of sample data to ensure the required accuracy of identification and approximation of these distribution laws. Specific analytical dependences and graphs are given for determining such distribution laws as simple laws (normal, uniform, triangular) probability distributions, oneparameter laws (gamma distributions, lognormal distribution, Weibull distribution), twoparameter beta distribution.

Keywords

Probability distribution density, Sample data volume, Modeling of distribution laws