Goodness-of-fit tests for conditional models under censoring and truncation

Cao, Ricardo; González–Manteiga, Wenceslao

doi:10.1016/j.jeconom.2007.08.011

Cited by 8 publications

(6 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…wherep is a nonparametric estimator with bandwidth h,θ estimates θ, and π(•) is a given density function. For practical purposes we recommend an empirical versionT n of the special case of T n where π(•) equals the covariate density f (•) (Cao and González-Manteiga, 2008),…”

Section: Model and Theoretical Resultsmentioning

confidence: 99%

Goodness-of-fit tests for the cure rate in a mixture cure model

Müller

Keilegom

2018

Biometrika

View full text Add to dashboard Cite

We consider models for time-to-event data that allow that an event, e.g., a relapse of a disease, never occurs for a certain percentage p of the population, called the cure rate. We suppose that these data are subject to random right censoring and we model the data using a mixture cure model, in which the survival function of the uncured subjects is left unspecified. The aim is to test whether the cure rate p, as a function of the covariates, satisfies a certain parametric model. To do so, we propose a test statistic that is inspired by a goodness-of-fit test for a regression function by Härdle and Mammen. We show that the statistic is asymptotically normally distributed under the null hypothesis that the model is correctly specified and under local alternatives. A bootstrap procedure is proposed to implement the test. The good performance of the approach is confirmed with simulations. For illustration we apply the test to data on the times between first and second birth.

show abstract

Section: Model and Theoretical Resultsmentioning

confidence: 99%

Goodness-of-fit tests for the cure rate in a mixture cure model

Müller

Keilegom

2018

Biometrika

View full text Add to dashboard Cite

show abstract

“…For instance, it is noted that a non-parametric estimator for the regression function in the double-truncated setting is already available. Following the ideas in Cao and González-Manteiga (2008), the design of a parametric estimator allowing an i.i.d. representation would enable the construction of a GoF test.…”

Section: (D) Non-and Semiparametric Hypothesesmentioning

confidence: 99%

Rejoinder on: An updated review of Goodness-of-Fit tests for regression models

González–Manteiga

Crujeiras

2013

TEST

Self Cite

View full text Add to dashboard Cite

First for all, we would like to thank the discussants for reading our paper and for taking time to prepare such interesting and valuable contributions. The feeling, after revising the discussions and going back to the original version of the paper, is that there is still too much to say about Goodness-of-Fit (GoF) tests for regression models and the discussants have given a good proof of this. Although they qualify the review as thorough and nearly exhaustive, it is clear after their comments there are quite a few issues and open problems that were left untreated.We have tried to organize this rejoinder according to the different topics that have been brought up in the discussions, instead of answering each discussion separately. Bearing this in mind, some brief comments will be made on (a) tests based on the error distribution; (b) tests with dependence between error and covariate; (c) bandwidth selection and calibration; (d) non-and semiparametric hypotheses; (e) tests in survival analysis and (f) some further topics.(a) Tests based on the error distribution In Sect. 2.4, concerned with this topic, two main methodological approaches are presented. The first one, collected in the paper by Van Keilegom et al. (2008), is based on the empirical distribution of the residuals, whereas Huskova and Meintanis (2009) considered an alternative route This rejoinder refers to the comments available at

show abstract

“…In comparison with the former, the latter approach has the advantage of being easier to compute, since it does not require any iterative computation. The method proposed in this paper is inspired by Cao and González-Manteiga (2008), who study a least squares procedure for the case where the coefficients are considered as being time-independent.…”

Section: Introductionmentioning

confidence: 99%