Testing Censoring Point Independence
نویسنده
چکیده
Identification in censored regression analysis and hazard models of duration outcomes relies on the condition that censoring points are independent of latent outcomes, an assumption which may be questionable in many settings. This note proposes a test for this assumption based on a Kolmogorov-Smirnov-like test statistic comparing two different nonparametric estimators for the latent outcome cdf: the empirical cdf conditional on the censoring point exceeding (for right-censored data) the cdf evaluation point, and the Kaplan-Meier estimator. The test is consistent and has power against a wide variety of alternatives. Applying the test to data from several studies of censored outcomes suggests the assumption is frequently suspect. keywords: survival analysis, duration data, Kaplan-Meier estimator, censored regression, hazard models
منابع مشابه
CONSTANT STRESS ACCELERATED LIFE TESTING DESIGNWITH TYPE-II CENSORING SCHEME FOR PARETO DISTRIBUTION USING GEOMETRIC PROCESS
In many of the studies concerning Accelerated life testing (ALT), the log linear function between life and stress which is just a simple re-parameterization of the original parameter of the life distribution is used to obtain the estimates of original parameters but from the statistical point of view, it is preferable to work with the original parameters instead of developing inferences for the...
متن کاملModified Progressive Type-II Censoring Procedure in Life-Testing under the Weibull Model
In this paper we introduce a new scheme of censoring and study it under the Weibull distribution. This scheme is a mixture of progressive Type II censoring and self relocating design which was first introduced by Srivastava [8]. We show the superiority of this censoring scheme (PSRD) relative to the classical schemes with respect to “asymptotic variance”. Comparisons are also made with respect ...
متن کاملExact hypothesis testing and confidence interval for mean of the exponential distribution under Type-I progressive hybrid censoring
Censored samples are discussed in experiments of life-testing; i.e. whenever the experimenter does not observe the failure times of all units placed on a life test. In recent years, inference based on censored sampling is considered, so that about the parameters of various distributions such as normal, exponential, gamma, Rayleigh, Weibull, log normal, inverse Gaussian, ...
متن کاملTesting the assumptions for the analysis of survival data arising from a prevalent cohort study with follow-up.
In a prevalent cohort study with follow-up subjects identified as prevalent cases are followed until failure (defined suitably) or censoring. When the dates of the initiating events of these prevalent cases are ascertainable, each observed datum point consists of a backward recurrence time and a possibly censored forward recurrence time. Their sum is well known to be the left truncated lifetime...
متن کاملComparison between constant-stress testing and step-stress testing under Type-I censoring Comparison between Constant-stress Testing and Step-stress Testing under Type-I Censoring
By running the life tests at higher stress levels than normal operating conditions, accelerated life testing quickly yields information on the lifetime distribution of a test unit. The lifetime at the design stress is then estimated through extrapolation using a regression model. In constant-stress testing, a unit is tested at a fixed stress level until failure or the termination time point of ...
متن کامل