Application of a Nonparametric Pattern Recognition Algorithm to the Problem of Testing the Hypothesis of the Independence of Variables of Multi-Valued Functions

The problem of hypothesis testing for the independence of two-dimensional random variables in the analysis of variables of multi-valued functions is considered. To solve it, we used a technique based on a nonparametric kernel-type pattern recognition algorithm corresponding to the maximum likelihood...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Measurement techniques Jg. 65; H. 1; S. 17 - 23
Hauptverfasser: Lapko, A. V., Lapko, V. A., Bakhtina, A. V.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Springer US 01.04.2022
Springer
Springer Nature B.V
Schlagworte:
Algorithms
Analytical Chemistry
Characterization and Evaluation of Materials
Hypotheses
Measurement Science and Instrumentation
Nonparametric statistics
Object recognition
Pattern recognition
Physical Chemistry
Physics
Physics and Astronomy
Random noise
Random variables
Remote sensing
Two dimensional analysis
maximum likelihood test
kernel estimation of probability density
hypothesis testing
multi-valued functional dependencies
dependent random variables
two-dimensional random variables
independent random variables
pattern recognition algorithm
ISSN:0543-1972, 1573-8906
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The problem of hypothesis testing for the independence of two-dimensional random variables in the analysis of variables of multi-valued functions is considered. To solve it, we used a technique based on a nonparametric kernel-type pattern recognition algorithm corresponding to the maximum likelihood criterion. The technique made it possible to bypass the problem of decomposing the random variable domain of values into intervals. Based on the results of computational experiments, the effectiveness of the applied technique was estimated depending on the type of multi-valued functions, the level of random noise and the amount of initial statistical data. The results obtained are relevant for solving the problem of detecting natural and technical objects from remote sensing data.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0543-1972
1573-8906
DOI:10.1007/s11018-022-02043-2