Detail výsledku
T-test alternative for multiply left-censored samples from Weibull distribution
Michálek Jaroslav, doc. RNDr., CSc., FSI (FSI)
Left-censored data with one or more detection limits occur frequently in many application areas. This paper suggests a computational procedure for calculation of maximum likelihood estimates of parameters for type I multiply left-censored Weibull samples. Moreover, using the asymptotic properties of maximum likelihood estimates, an alternative to classic t-test that can be used for comparison of expected values of two independent type I multiply left-censored Weibull samples is proposed. The power functions of both tests are compared by simulations. The methods derived in this paper can be used in real environmental or chemical data analysis.
Fisher information matrix, maximum likelihood, power function of test, t-test
@inproceedings{BUT115030,
author="Michal {Fusek} and Jaroslav {Michálek}",
title="T-test alternative for multiply left-censored samples from Weibull distribution",
booktitle="21st International Conference on Soft Computing - MENDEL 2015",
year="2015",
journal="Mendel Journal series",
pages="169--174",
issn="1803-3814"
}