Effect of Informative Censoring on the Power of Some Log-Rank Type Tests.

The survival analysis is a set of techniques for analyzing the elapsed time until the occurrence of an event of interest, also called lifetime, which can be completely or partially observed. In practice, a common problem is to compare the survival experiences of at least two groups...

Full description

Bibliographic Details
Main Authors: Brun González, L. P., Salazar Uribe, J. C.
Format: Online
Language:spa
Published: Universidad Pedagógica y Tecnológica de Colombia 2016
Subjects:
Online Access:https://revistas.uptc.edu.co/index.php/ciencia_en_desarrollo/article/view/4230
Description
Summary:The survival analysis is a set of techniques for analyzing the elapsed time until the occurrence of an event of interest, also called lifetime, which can be completely or partially observed. In practice, a common problem is to compare the survival experiences of at least two groups under study. The analysis strategy most common is to use Log-Rank tests, but these do not distinguish whether the mechanism of censoring it is informative or not. In this work it was evaluated, under different simulation scenarios, the estimated power of some Log-Rank tests when informative censoring is presented. We found evidence favoring the Wilcoxon test.