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

Adjusted variable plots for Cox's proportional hazards regression model

C B Hall1, S L Zeger, K J Bandeen-Roche

  • 1Department of Statistics, University of Connecticut, Storrs 06269, USA. chall@stat.uconn.edu

Lifetime Data Analysis
|January 1, 1996
PubMed
Summary

This study introduces novel adjusted variable plots for Cox proportional hazards models, enhancing outlier detection and model fit evaluation in survival data analysis. These graphical tools improve the interpretation of complex survival models.

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

The Chemical Treatment of Diseases.

The Chicago medical journal·2023
Same author

Centrally Acting ACE Inhibitor Use and Physical Performance in Older Adults.

The Journal of frailty & aging·2023
Same author

World Trade Center exposure, post-traumatic stress disorder, and subjective cognitive concerns in a cohort of rescue/recovery workers.

Acta psychiatrica Scandinavica·2019
Same author

Relationship between body composition and the components of somatotype.

American journal of human biology : the official journal of the Human Biology Council·2017
Same author

Statin use and neurologic morbidity after coronary artery bypass grafting: A cohort study.

Neurology·2009
Same author

Cognitive activities delay onset of memory decline in persons who develop dementia.

Neurology·2009

Area of Science:

  • Biostatistics
  • Survival Analysis
  • Statistical Modeling

Background:

  • Adjusted variable plots are valuable for assessing linear regression models, aiding in outlier identification and model fit evaluation.
  • Survival data analysis often involves complex models like Cox's proportional hazards model, necessitating robust diagnostic tools.

Purpose of the Study:

  • To extend the utility of adjusted variable plots to Cox's proportional hazards model for censored survival data.
  • To develop and evaluate new graphical methods for diagnosing Cox models.

Main Methods:

  • Proposed three novel adjusted variable plots: Risk Level Adjusted Variable (RLAV), Subject Level Adjusted Variable (SLAV), and Event Level Adjusted Variable (ELAV).
  • RLAV plots display all observations within each risk set.

Related Experiment Videos

  • SLAV and ELAV plots aggregate points from RLAV plots to represent subjects or event risk sets, respectively. Regression coefficients and standard errors are derived via linear regression through the origin.

    • Main Results:

    • The proposed plots provide visual diagnostics for Cox proportional hazards models.
    • Demonstrated the application of these plots through a reanalysis of a multiple myeloma patient dataset.
    • The methods facilitate qualitative evaluation of model fit and outlier detection in survival analysis.

    Conclusions:

    • The extended adjusted variable plots offer effective graphical tools for the assessment of Cox proportional hazards models.
    • These methods enhance the interpretability and diagnostic capabilities for survival data analysis.
    • The plots are particularly useful for identifying influential observations and assessing overall model performance.