ISSN:
1572-9249
Keywords:
right-censored data
;
reporting delays
;
influence curve
;
Kaplan-Meier estimator
Source:
Springer Online Journal Archives 1860-2000
Topics:
Mathematics
Notes:
Abstract In disease registries there can be a delay between death of asubject and the reporting of this death to the data analyst.If researchers use the Kaplan-Meier estimator and implicitlyassumed that subjects who have yet to have death reported arestill alive, i.e. are censored at the time of analysis, the Kaplan-Meierestimator is typically inconsistent. Assuming censoring is independentof failure, we provide a simple estimator that is consistentand asymptotically efficient. We also provide estimates of theasymptotic variance of our estimator and simulations that demonstratethe favorable performance of these estimators. Finally, we demonstrateour methods by analyzing AIDS survival data. This analysis underscoresthe pitfalls of not accounting for delay when estimating thesurvival distribution and suggests a significant reduction inbias by using our estimator.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1023/A:1009689625311
Permalink
