Statistical power of goodness-of-fit tests based on the empirical distribution function for type-I right-censored data

Bispo, Regina; Marques, Tiago A.; Pestana, Dinis

http://hdl.handle.net/10400.12/1809

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
JSCS iFirst 2011 1–9.pdf		336.6 KB	Adobe PDF	Download

Send Feedback

Authors

Bispo, Regina

Marques, Tiago A.

Pestana, Dinis

Abstract(s)

In this study, the power of common goodness-of-fit (GoF) statistics based on the empirical distribution function (EDF) was simulated for single type-I right-censored data. The statistical power of the Kolmogorov–Smirnov, Cramér–von Mises and Anderson–Darling statistics was investigated by varying the null and the alternative distributions, the sample size, the degree of censoring and the significance level. The exponential,Weibull, log-logistic and log-normal lifetime distributions were considered as they are among the most frequently distributions used when modelling censored data. We conclude by giving some general recommendations for testing the distributional assumption of parametric survival models in homogeneous populations when using EDF-based GoF statistics.