Pseudo-partial likelihood estimators for the Cox regression model with missing covariates.

Biometrika

Department of Psychiatry , Mount Sinai School of Medicine, New York, New York 10029 , U.S.A.

Published: September 2009

By embedding the missing covariate data into a left-truncated and right-censored survival model, we propose a new class of weighted estimating functions for the Cox regression model with missing covariates. The resulting estimators, called the pseudo-partial likelihood estimators, are shown to be consistent and asymptotically normal. A simulation study demonstrates that, compared with the popular inverse-probability weighted estimators, the new estimators perform better when the observation probability is small and improve efficiency of estimating the missing covariate effects. Application to a practical example is reported.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3741327PMC
http://dx.doi.org/10.1093/biomet/asp027DOI Listing

Publication Analysis

Top Keywords

pseudo-partial likelihood
8
likelihood estimators
8
cox regression
8
regression model
8
model missing
8
missing covariates
8
missing covariate
8
estimators
5
estimators cox
4
missing
4

Similar Publications

Nested case-control sampling without replacement.

Lifetime Data Anal

October 2024

University of Maryland, College Park, Maryland, United States.

Nested case-control design (NCC) is a cost-effective outcome-dependent design in epidemiology that collects all cases and a fixed number of controls at the time of case diagnosis from a large cohort. Due to inefficiency relative to full cohort studies, previous research developed various estimation methodologies but changing designs in the formulation of risk sets was considered only in view of potential bias in the partial likelihood estimation. In this paper, we study a modified design that excludes previously selected controls from risk sets in view of efficiency improvement as well as bias.

View Article and Find Full Text PDF
Article Synopsis
  • The study focuses on how well a biomarker predicts time-to-event outcomes in different patient subgroups, emphasizing that traditional measures like AUC may not capture important differences.
  • It introduces the partial time-dependent AUC (PAUC) as a more relevant metric for evaluating biomarker performance in screening and diagnostic tests, along with a specialized regression model for this purpose.
  • The researchers validate their model using simulation studies and apply it to Alzheimer's Disease data to explore how various patient characteristics affect the biomarkers’ predictive abilities for early diagnosis.
View Article and Find Full Text PDF

Accurate discrimination has been the central goal in identifying biomarkers for monitoring disease progression and early detection. Acknowledging the fact that discrimination accuracy of biomarkers for a time-to-event outcome often changes over time, local measures such as the time-dependent receiver operating characteristic curve and its area under the curve (AUC) are used to assess time-dependent predictive discrimination. However, such measures do not address subject heterogeneity, although the impact of covariates including demographics, disease-related characteristics, and other clinical information on the discriminatory performance of biomarkers needs to be investigated before their clinical use.

View Article and Find Full Text PDF

Infectious pathogens and risk of esophageal, gastric and duodenal cancers and ulcers in China: A case-cohort study.

Int J Cancer

April 2024

Clinical Trial Service Unit & Epidemiological Studies Unit (CTSU), Nuffield Department of Population Health, University of Oxford, Oxford, UK.

Infection by certain pathogens is associated with cancer development. We conducted a case-cohort study of ~2500 incident cases of esophageal, gastric and duodenal cancer, and gastric and duodenal ulcer and a randomly selected subcohort of ~2000 individuals within the China Kadoorie Biobank study of >0.5 million adults.

View Article and Find Full Text PDF

Risk prediction models for survival outcomes are widely applied in medical research to predict future risk for the occurrence of the event. In many clinical studies, the biomarker data are measured repeatedly over time. To facilitate timely disease prognosis and decision making, many dynamic prediction models have been developed and generate predictions on a real-time basis.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!