Approximation algorithms for the maximum likelihood estimation of the parameters of the trapezoidal distribution
https://doi.org/10.29235/1561-2430-2026-62-2-143-148
Abstract
We considered the problem of estimation of the parameters of the trapezoidal distribution. We built a confidence interval. This confidence interval with high probability contains the value of the random variable with a trapezoidal distribution. We further developed two approximation algorithms for the problem that are based on the maximum likelihood principle. Algorithm uses the enumeration principle of the values of the parameters. The trapezoidal distribution may be used instead of normal distribution when there are upper and lower bounds on the value of the random variable.
Supporting agencies: The research is partially supported by the Belarusian Republican Foundation for Fundamental Research (project no. Ф26КИ-009).
About the Author
M. S. Barketau
United Institute of Informatics Problems of the National Academy of Sciences of Belarus
Belarus
Belarus
Maksim S. Barketau – Ph. D. (Physics and Mathematics), Associate Professor, Senior Researcher
6, Surganov Str., 220012, Minsk
References
1. Kacker R. N., Lawrence J. F. Trapezoidal and triangular distributions for Type B evaluation of standard uncertainty. Metrologia, 2007, vol. 44, no. 2, pp. 117–127. https://doi.org/10.1088/0026-1394/44/2/003
2. Dorp van J. R., Kotz S. Generalized trapezoidal distributions. Metrika, 2003, vol. 58, pp. 85–97. https://doi.org/10.1007/s001840200230
3. Zygmunt L. W. Effective measurand estimators for samples of trapezoidal PDFs. Journal of Automation, Mobile Robotics and Intelligent Systems, 2012, vol. 6, no. 1, pp. 8–14.
4. Borovkov A. A. Mathematical Statistics. Saint Petersburg, Lan’ Publ., 2010. 704 p. (in Russian).
5. Borovkov A. A. Probability Theory. Moscow, Editorial URSS Publ., 1999. 472 p. (in Russian).
6. Wainwright M. J. High-Dimensional Statistics: A Non-Asymptotic Viewpoint. Cambridge University Press, 2019. 571 p. https://doi.org/10.1017/9781108627771
Review
Views:
48
JATS XML
Email this article 
