NONPARAMETRIC ALGORITHMS OF PATTERN RECOGNITION IN THE PROBLEM OF HYPOTHESIS TESTING ON DISTRIBUTIONS OF RANDOM VARIABLES
Institute of Computational Modeling, Siberian Branch of RAS; Professor
V. A. Lapko
M. F. Reshetnev Siberian State Aerospace University, Department of Space Resources and Technologies; Professor
Abstract. A new method of hypothesis testing on identity of distributions of random variables is proposed. The method is based on application of nonparametric algorithms of pattern recognition and collective estimation principles. Results obtained with the method are compared with Kolmogorov — Smirnov criterion.
Keywords: nonparametric statistics, pattern recognition, hypothesis testing, distributions of random variables.