Statistical Contributions to Reliability Engineering,
Abstract
In chapter 1 parametric and distribution-free techniques are developed to treat simultaneously any arbitrary set of multi-censored failure data and some of their practical implications are briefly discussed. In chapter 2 some general properties of probability limits on the ith order statistic (1 < or = i < or = n) from any continuous failure law are briefly investigated. In chapter 3 interrelationships between 100 alpha percent (0 < alpha < 1) probability limits on the ith order statistics (1 < or = i < or = n) from various failure distributions are established. This yields suitable transformations indicating interesting connections between various distributions including Weibull, Burr and others. In chapter 4 some distribution-free results are derived which are useful within the context of certain life testing experiments. In chapter 5 a likelihood ratio test, based on right-censored samples, is derived for testing the equality of location parameters of two exponential distributions when their common scale parameter is unknown. This has the well-known F distribution under the null hypothesis. The power functions of the likelihood ratio test are also derived and it is shown that this test is unbiased. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1972
- Accession Number
- AD0751261
Entities
People
- Satya D. Dubey
Organizations
- New York University