Testing for Intervals of Increased Mortality.
Abstract
To test the hypothesis of constant hazard rate against the alternative that there is a time interval in which that hazard rate is increased, a new statistic is proposed, in addition to some previously known statistics. A procedure is described and illustrated on examples, which uses computers and makes it possible to use these statistics in practice, although their sampling distributions are either not known or not numerically tabulated.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 30, 1974
- Accession Number
- ADA001799
Entities
People
- Z. W. Birnbaum
Organizations
- University of Washington