The conditional maximum likelihood estimation for the Cox-Aalen model with doubly truncated data

Doubly truncated (DT) data occur when event times are observed only if they fall within subject-specific, possibly random, intervals. In this article, we consider conditional maximum likelihood estimation for the regression parameters of the Cox-Aalen model with DT data. Based on gradient projection...

Full description

Saved in:
Bibliographic Details
Published inStatistics (Berlin, DDR) Vol. 59; no. 1; pp. 228 - 245
Main Authors Su, Chun-Lung, Shen, Pao-sheng
Format Journal Article
LanguageEnglish
Published Abingdon Taylor & Francis 02.01.2025
Taylor & Francis Ltd
Subjects
additive hazard regression
Algorithms
conditional likelihood
Cox model
double truncation
EM algorithm
Maximum likelihood estimation
Maximum likelihood estimators
Parameter estimation
Primary: 62N99
Regression models
Secondary: 62N02
Simulation
Online AccessGet full text

Cover

More Information
Summary:Doubly truncated (DT) data occur when event times are observed only if they fall within subject-specific, possibly random, intervals. In this article, we consider conditional maximum likelihood estimation for the regression parameters of the Cox-Aalen model with DT data. Based on gradient projection method (GPM), we propose computational algorithms for obtaining the conditional maximum likelihood estimator (cMLE). The proposed cMLE is shown to be consistent and asymptotically normal. Simulation studies show that the cMLE performs well in finite samples.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0233-1888
1029-4910
DOI:10.1080/02331888.2024.2431737