Weighted Least Squares Method for the Accelerated Failure Time Model with Auxiliary Covariates

This paper deals with the analysis of accelerated failure time model when the primary covariate is subject to missing. We assume that the true covariate is measured precisely on a randomly chosen validation set, whereas auxiliary information for primary covariate is available to all study subjects....

Full description

Saved in:
Bibliographic Details
Published inActa mathematica Sinica. English series Vol. 35; no. 7; pp. 1163 - 1178
Main Authors Jin, Ling Hui, Liu, Yan Yan, Wu, Lang
Format Journal Article
LanguageEnglish
Published Beijing Institute of Mathematics, Chinese Academy of Sciences and Chinese Mathematical Society 01.07.2019
Springer Nature B.V
Subjects
Asymptotic properties
Computer simulation
Failure analysis
Failure times
Least squares method
Mathematics
Mathematics and Statistics
Simulation
accelerated failure time model
Survival analysis
62N01
62N02
validation sample
auxiliary covariates
weighted least squares estimation
Online AccessGet full text

Cover

More Information
Summary:This paper deals with the analysis of accelerated failure time model when the primary covariate is subject to missing. We assume that the true covariate is measured precisely on a randomly chosen validation set, whereas auxiliary information for primary covariate is available to all study subjects. The asymptotic properties for the proposed estimator are developed and the simulation studies show that the efficiency gain is remarkable compared to the method using only the validation sample. A real example is also provided as an illustration.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1439-8516
1439-7617
DOI:10.1007/s10114-019-8232-9