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

Comparing non-nested regression models

P Royston1, S G Thompson

  • 1Medical Statistics Unit, Royal Postgraduate Medical School, London, England.

Biometrics
|March 1, 1995
PubMed
Summary
This summary is machine-generated.

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

Value of risk scores in the decision to palliate patients with ruptured abdominal aortic aneurysm.

The British journal of surgery·2018
Same author

Lessons learned about prevalence and growth rates of abdominal aortic aneurysms from a 25-year ultrasound population screening programme.

The British journal of surgery·2017
Same author

Sex differences in mortality after abdominal aortic aneurysm repair in the UK.

The British journal of surgery·2017
Same author

Meta-analysis of the current prevalence of screen-detected abdominal aortic aneurysm in women.

The British journal of surgery·2016
Same author

Cost-effectiveness of the National Health Service Abdominal Aortic Aneurysm Screening Programme in England.

The British journal of surgery·2014
Same author

Long-term cost-effectiveness analysis of endovascular versus open repair for abdominal aortic aneurysm based on four randomized clinical trials.

The British journal of surgery·2014
Same journal

Fast penalized generalized estimating equations for large longitudinal functional datasets.

Biometrics·2026
Same journal

Causally-interpretable random-effects meta-analysis.

Biometrics·2026
Same journal

Statistical inference for mean function of partially observed functional time series.

Biometrics·2026
Same journal

Subgroup identification via Interaction Tree and Mixed Model for Repeated Measures with application to Alzheimer's disease.

Biometrics·2026
Same journal

Finite mixtures of linear quantile regressions with concomitant variables: a solution to endogeneity in longitudinal data modeling.

Biometrics·2026
Same journal

Discussion on "INTACT: a method for integration of longitudinal physical activity data from multiple sources" by Jingru Zhang, Erjia Cui, Hongzhe Li, and Haochang Shou.

Biometrics·2026
See all related articles

This study introduces a novel statistical method for comparing non-nested regression models. The approach uses an artificial supermodel to test model fit, simplifying complex comparisons for researchers.

Area of Science:

  • Statistics
  • Econometrics
  • Regression Analysis

Background:

  • Comparing non-nested statistical models is challenging.
  • Existing methods may lack generality or computational simplicity.

Purpose of the Study:

  • To develop a robust method for comparing the fits of two non-nested models.
  • To extend existing model comparison techniques for broader applicability.

Main Methods:

  • A novel approach based on Davidson and MacKinnon (1981) is developed.
  • An artificial 'supermodel' is constructed, encompassing both individual models.
  • Fitted values from individual models are used in an approximate supermodel for estimation and hypothesis testing.

Main Results:

Related Experiment Videos

  • The proposed test simplifies to the standard F test for nested linear models.
  • Calculations primarily involve linear regression through the origin.
  • The method is extended to address bias in maximum likelihood estimation.

    • Conclusions:

    • The developed method provides a computationally straightforward way to compare non-nested regression models.
    • The technique is applicable to both linear and nonlinear regression with normal errors.
    • The study illustrates the method's utility with two real-world datasets.