Preview

Dependability

Advanced search

ABOUT CHECKING OF RANDOMNESS OF DATA WITH TIES

https://doi.org/10.21683/1729-2646-2015-0-2-39-49

Abstract

The paper considers research of various nonparametric methods of checking a hypothesis of randomness of sampled data in a situation when these data contain repeating (tied) observations. The repeating observations named ties lead to ambiguous ranking. The arithmetic-mean rank is usually assigned to elements of the tied group. As a result, statistical decision rules change as the criterion statistics changes its distribution. This paper is devoted to research of capacity of various criteria at presence of ties.

About the Authors

V. S. Ninchuk
Obninsk Institute of Atomic Energy, National Research Nuclear University MIFI
Russian Federation
Post graduate student of chair of Automated Control Systems


V. A. Chepurko
Obninsk Institute of Atomic Energy, National Research Nuclear University MIFI
Russian Federation
PhD, associate professor of chair of automated control systems


References

1. Hollender М., Wolf D. Non-parametric methods of statistics. M.: Finance and statistics, 1983.-518 p.

2. Antonov A.V., Chepurko V.A. Integral test for checking trend hypothesis. Dependability, 2006. No, 2. Pp. 17-27.

3. Gaek Ia., Shidak Z. The theory of rank tests. M.: Science, 1971.-376 p.

4. Burtaev J.F., Ostreykovsky V.A. Statistical analysis of dependability of objects with limited information. M.: Energoatom izdat, 1995,-240 p.

5. Kendall M. Rank correlations.Foreign statistical studies. М., «Statistics», 1975.216 p.

6. Van der Varden. Mathematical statistics. M.: Foreign literature edition, 1960.435 p.


Review

For citations:


Ninchuk V.S., Chepurko V.A. ABOUT CHECKING OF RANDOMNESS OF DATA WITH TIES. Dependability. 2015;(2):39-49. https://doi.org/10.21683/1729-2646-2015-0-2-39-49

Views: 649


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.


ISSN 1729-2646 (Print)
ISSN 2500-3909 (Online)