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

Jackknife bias reduction for polychotomous logistic regression

S B Bull1, C M Greenwood, W W Hauck

  • 1Samuel Lunenfeld Research Institute, University of Toronto, Ontario, Canada.

Statistics in Medicine
|March 15, 1997
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

HostSeq: a Canadian whole genome sequencing and clinical data resource.

BMC genomic data·2023
Same author

Transfemoral catheterization of the cavernous sinus outflow with comparison of "central" and "peripheral" gonadotropin patterns in bonnet monkeys.

American journal of primatology·2020
Same author

Rare susceptibility variants for bipolar disorder suggest a role for G protein-coupled receptors.

Molecular psychiatry·2017
Same author

Using cancer registry data: agreement in cause-of-death data between the Ontario Cancer Registry and a longitudinal study of breast cancer patients.

Chronic diseases in Canada·2009
Same author

Molecular and epidemiologic study of multiresistant Serratia marcescens Infections in a spinal cord injury rehabilitation unit.

Infection control and hospital epidemiology·2009
Same author

Association of apolipoprotein E polymorphisms and dietary factors in colorectal cancer.

British journal of cancer·2009

Bias-reduced estimates using jackknife or Taylor series methods are useful in logistic regression with sufficient data. However, these methods are not recommended for smaller sample sizes or when dealing with complex covariate structures.

Area of Science:

  • Statistics
  • Biostatistics
  • Regression Analysis

Background:

  • Maximum likelihood estimates (MLEs) can be misleading in finite samples.
  • Bias-reduced estimation methods offer potential improvements in bias and efficiency.
  • These methods have not been widely adopted in practical applications.

Purpose of the Study:

  • To detail bias-reduced estimation methods for polychotomous logistic regression.
  • To compare jackknife and Taylor series estimates via Monte Carlo simulation.
  • To investigate performance in moderate sample sizes with various covariate types.

Main Methods:

  • Jackknife methods (with and without full iteration).
  • Taylor series expansion of the log-likelihood.
  • Monte Carlo simulation comparing estimation methods.

Related Experiment Videos

Main Results:

  • Approximate two-step jackknife and Taylor series methods are useful when observations-to-parameters ratio > 15.
  • Two-step and fully iterated jackknife estimates are not recommended when this ratio < 20.
  • Performance issues are exacerbated by large effects, binary covariates, or multicollinearity.

Conclusions:

  • Bias-reduced methods can be effective in logistic regression under specific conditions.
  • Sample size and covariate characteristics are critical for method selection.
  • Careful consideration of these factors is needed for reliable estimation.