A note on the use of the shapiro- wilk test of exponentiality for complete samples
- 1 January 1989
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Theory and Methods
- Vol. 18 (5) , 1819-1830
- https://doi.org/10.1080/03610928908830003
Abstract
The consistency of the Shapiro-Wilk test of exponentialIty is examined. In testing repair-time data for exponentiality, the natural class of lognormal alternatives is shown to contain a subclass over which the test Is Inconsistent. A transformation of the data is recommended to alleviate the problem.Keywords
This publication has 6 references indexed in Scilit:
- The Shapiro-Wilk Test for Exponentiality Based on Censored DataJournal of the American Statistical Association, 1988
- Inferences on the lognormal mean for complete samplesCommunications in Statistics - Simulation and Computation, 1988
- On a geometrical derivation of the upper tail of the null distribution of w-exponentialCommunications in Statistics - Theory and Methods, 1985
- On the W Test for Exponentiality with Origin KnownTechnometrics, 1978
- An Analysis of Variance Test for the Exponential Distribution (Complete Samples)Technometrics, 1972
- Confidence Intervals for Linear Functions of the Normal Mean and VarianceThe Annals of Mathematical Statistics, 1971