A Note on Some Simple Influence Diagnostics in Accelerated Life Testing.
Abstract
Three methods for assessing influence in an accelerated life-testing model are considered; two different one-step approximations to the estimated parameter after case deletion and a method which treats extreme values at each design point is censored. These methods are compared using an example. Problems which occur when all observations at a design point are censored are discussed. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1984
- Accession Number
- ADA149593
Entities
People
- H. Schneider
- L. Weissfeld
- Raymond J. Carroll
Organizations
- University of North Carolina at Chapel Hill