A Power Analysis of Tests for Paired Lifetime Data

Author: Owen, William

Source: Lifetime Data Analysis, Volume 11, Number 2, June 2005 , pp. 233-243(11)

Publisher: Springer

Buy & download fulltext article:

Price: $47.00 plus tax (Refund Policy)

Abstract:

A power analysis is used to compare tests developed for a new lifetime model applicable for certain paired-data experiments. Two pivotal quantities are available for exact parametric testing of the equality of the marginal survival distributions. Here, the power of each test will be calculated to find the more powerful test. The analysis leads to an interesting scenario where the two power curves are quite similar and less powerful test may actually be preferred for pragmatic considerations.

Keywords: asymptotic relative efficiency; bivariate data; logistic distribution; simulation; survival analysis; Weibull distribution

Document Type: Research article

DOI: http://dx.doi.org/10.1007/s10985-004-0385-9

Affiliations: 1: Department of Mathematics and Computer Science, University of Richmond, Richmond, VA, 23173, USA, Email: wowen@richmond.edu

Publication date: 2005-06-01