Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Non-parametric estimation for baseline hazards function and covariate effects with time-dependent covariates.

Feng Gao1, Amita K Manatunga, Shande Chen

  • 1Division of Biostatistics, Washington University School of Medicine, Campus Box 8067, 660 S. Euclid Ave., St Louis, MO 63110, USA.

Statistics in Medicine
|May 11, 2006
PubMed
Summary

This study introduces a novel tree-type method for estimating hazard functions, improving upon Breslow's estimator by handling time-dependent covariates and exploring data structures without formal hypothesis testing.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Model-based clustering of multiple images incorporating covariates.

Statistical methods in medical research·2026
Same author

Incidence of hidradenitis suppurativa in transgender and cisgender individuals: A multicenter matched cohort study.

Journal of the American Academy of Dermatology·2026
Same author

Acne Incidence and Severity in Transgender Individuals.

JAMA dermatology·2026
Same author

Assessing intra- and inter-method agreement of functional data.

Statistical methods in medical research·2023
Same author

An integrative latent class model of heterogeneous data modalities for diagnosing kidney obstruction.

Biostatistics (Oxford, England)·2023
Same author

Special issue dedicated to David Oakes.

Lifetime data analysis·2022

Area of Science:

  • Biostatistics
  • Survival Analysis
  • Epidemiology

Background:

  • Estimating hazard functions is crucial in biomedical and epidemiologic studies.
  • Breslow's estimator for integrated baseline hazard requires correct covariate effect specification, which is challenging with time-dependent covariates.
  • Traditional proportional hazard models struggle with accurately specifying functional forms for time-dependent covariates.

Purpose of the Study:

  • To propose a complementary tree-type method for estimating hazard functions.
  • To simultaneously estimate baseline hazards and time-dependent covariate effects.
  • To explore potential data structures and provide an alternative to traditional methods.

Main Methods:

  • A tree-type method is proposed to approximate baseline hazards and covariate effects using step-functions.

Related Experiment Videos

  • An algorithm searches for jump points in time and covariate space based on log-likelihood improvement.
  • The method directly estimates the hazards function, unlike methods estimating integrated hazards.

    • Main Results:

    • The proposed method successfully estimates both baseline hazards and time-dependent covariate effects.
    • The method was applied to model withdrawal risk in an anti-depression clinical trial.
    • Simulation studies demonstrated the performance of the developed method.

    Conclusions:

    • The novel tree-type method offers a robust alternative for hazard function estimation, particularly with time-dependent covariates.
    • This approach enhances the analysis of complex survival data in biomedical and epidemiologic research.
    • The method's ability to explore data structures provides valuable insights beyond hypothesis testing.