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

Bias in risk estimation: application to Down's syndrome screening.

K L Williams1, A B J Nix

  • 1Department of Epidemiology, Statistics and Public Health, University of Wales College of Medicine, Heath Park, Cardiff CF14 4XN, UK.

Statistics in Medicine
|September 3, 2002
PubMed
Summary

This study quantifies bias in Gaussian density estimation and likelihood ratios, offering algebraic approximations and confidence intervals. Findings reveal bias patterns and variable risk quality in Down

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 genotype-phenotype landscape of familial amyotrophic lateral sclerosis in Australia.

Clinical genetics·2017
Same author

Precision of total and regional body fat estimates from dual-energy X-ray absorptiometer measurements.

The journal of nutrition, health & aging·2014
Same author

The mating system of the cellular slime mould Dictyostelium discoideum.

Current genetics·2013
Same author

Temperature-sensitive DNA repair mutations in the cellular slime mould Dictyostelium discoideum.

Current genetics·2013
Same author

Post-exercise cardiac troponin release is related to exercise training history.

International journal of sports medicine·2012
Same author

Exped 2010--the challenges of an expedition doctor.

Journal of the Royal Army Medical Corps·2011

Area of Science:

  • Statistics
  • Biostatistics
  • Medical Screening

Background:

  • Parametric estimation of Gaussian densities is fundamental in statistical analysis.
  • Bias in density estimation can impact downstream applications like likelihood ratio tests.
  • Accurate risk assessment is crucial in medical screening programs.

Purpose of the Study:

  • To derive and validate algebraic approximations for bias in univariate and bivariate Gaussian density estimation.
  • To quantify the induced bias when these densities are used for likelihood ratio determination.
  • To establish approximate confidence intervals for true density and likelihood ratio, and apply them to Down's syndrome screening.

Main Methods:

  • Derivation of algebraic approximations for bias in Gaussian density estimation.

Related Experiment Videos

  • Simulation exercises to verify the accuracy of derived approximations and confidence intervals.
  • Application of methods to estimate confidence intervals for posterior odds in Down's syndrome screening.

    • Main Results:

    • Algebraic expressions accurately predict relative biases in Gaussian density estimation.
    • Zero bias occurs at four Z-scores in univariate Gaussian density estimation; bias increases beyond two standard deviations.
    • Derived 95% confidence intervals show coverage between 94-97%, with variable widths for similar posterior odds in Down's syndrome screening.

    Conclusions:

    • The derived approximations effectively predict bias in Gaussian density estimation and likelihood ratios.
    • Confidence intervals provide a measure of uncertainty for estimated densities and likelihood ratios.
    • The application to Down's syndrome screening highlights that risks of seemingly equal magnitude can have different levels of certainty.