L-Moments Method: Introducing Novel Approach of Mixture Parameters Estimation and Its Implementation in Python Library Pysatl-Mpest

This paper addresses the problem of estimating the parameters of a mixture of distributions. One of the most popular ways to solve this problem is the Expectation Maximization algorithm. We introduce a novel alternative method for estimating the parameters of a mixture of distributions, based on numerical characteristics of distribution known as L-moments. Experimental results demonstrate significant speedup and comparable accuracy compared to classical methods.