scientific article; zbMATH DE number 3926043
[For the entire collection see Zbl 0564.00017.] Let \(X_ 1,X_ 2,...,X_ n\) be independent identically distributed random variables with a common continuous life distribution F. Let \(Y_ 1,Y_ 2,...,Y_ n\) be independent identically distributed random censoring times with a common continuous life distribution H. Consider the hypotheses \(H_ 0:\) F is an exponential distribution and \(H_ 1:\) F is HNBUE (harmonic new better than used in expectation) but not exponential. The second author [Scand. J. Stat., Theory Appl. 10, 65-75 (1983; Zbl 0522.62081)] introduced some test statistics for testing \(H_ 0\) versus \(H_ 1\) in the uncensored case. In the paper under review, these statistics are modified to the censored case. Computation formulas and asymptotic distributions are given. The efficiency loss due to censoring is given in a series of numerical tables. Finally, in a simulation study the authors investigate the accuracy of the normal approximation for the test statistics.
- scientific article; zbMATH DE number 4143256
- On testing hypotheses with doubly censored data
- A quick omnibus test for the proportional hazards model of random censorship
- Tests for increasing failure rate with randomly censored data
- Hypothesis testing with paired partly interval censored data
- Homogeneity tests for related samples under censoring
- Testing whether survival function is harmonic new better than used in expectation
- A note on some test statistics against HNBUE
- A family of test statistics for detecting monotone mean residual life
- Modeling of discrete HNBUE class with hypothesis testing application
- Testing whether new better than used in expectation for random censorship for random censorship
- An exact test for hnbue class of life distributions
- scientific article; zbMATH DE number 1057979 (Why is no real title available?)
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3700650)