Result Details

On testing reduction of a left-censored Weibull distribution to an exponential submodel

FUSEK, M. On testing reduction of a left-censored Weibull distribution to an exponential submodel. Mendel Journal series, 2017, vol. 23, no. 1, p. 179-184. ISSN: 1803-3814.
Type
journal article
Language
English
Authors
Abstract

When analyzing environmental or chemical data, it is often necessary to deal with left-censored observations. Since the distribution of the observed variable is often asymmetric, the exponential or the Weibull distribution can be used. This paper summarizes statistical model of a multiply left-censored Weibull distribution, and estimation of its parameters and their variances using the expected Fisher information matrix. Since in many situations the Weibull distribution is unnecessarily complicated for data modelling, statistical tests (the Lagrange multiplier test, the likelihood ratio test, the Wald test) for assessing suitability of replacement of the censored Weibull distribution with the exponential submodel are introduced and their power functions are analyzed using simulations.

Keywords

Asymptotic tests, multiply left-censored data, Fisher information matrix, maximum likelihood

Published
2017
Pages
179–184
Journal
Mendel Journal series, vol. 23, no. 1, ISSN 1803-3814
Book
23rd International Conference on Soft Computing - MENDEL 2017
EID Scopus
BibTeX
@article{BUT137290,
  author="Michal {Fusek}",
  title="On testing reduction of a left-censored Weibull distribution to an exponential submodel",
  journal="Mendel Journal series",
  year="2017",
  volume="23",
  number="1",
  pages="179--184",
  issn="1803-3814"
}
Departments
Back to top