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

Where is the likelihood ratio test powerful for detecting two component normal mixtures?

N R Mendell1, S J Finch, H C Thode

  • 1Department of Applied Mathematics and Statistics, State University of New York at Stony Brook 11794.

Biometrics
|September 1, 1993
PubMed
Summary

This study compares statistical tests for detecting mixtures of normal distributions. The Engelman-Hartigan test is best for moderate mixing proportions, while skewness and likelihood ratio tests perform well in other scenarios.

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

Initial chest radiograph scores inform COVID-19 status, intensive care unit admission and need for mechanical ventilation.

Clinical radiology·2021
Same author

Missed myocardial infarctions in ED patients prospectively categorized as low risk by established risk scores.

The American journal of emergency medicine·2017
Same author

Rocks, soils, and water quality. Relationships and implications for effects of acid precipitation on surface water in the Northeastern United States.

Environmental science & technology·2012
Same author

The power and robustness of maximum LOD score statistics.

Annals of human genetics·2008
Same author

Tradeoff between no-call reduction in genotyping error rate and loss of sample size for genetic case/control association studies.

Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing·2004
Same author

Errors and linkage disequilibrium interact multiplicatively when computing sample sizes for genetic case-control association studies.

Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing·2003

Area of Science:

  • Statistics
  • Statistical modeling
  • Hypothesis testing

Background:

  • Detecting mixtures of normal distributions is crucial in various scientific fields.
  • Understanding the performance of different statistical tests under varying mixture proportions is essential for accurate data analysis.

Purpose of the Study:

  • To compare the power of various statistical tests in detecting a mixture of two normal distributions with different means and equal variances.
  • To identify the most effective tests across the full range of mixing proportions (pi).

Main Methods:

  • Evaluated the power of 16 statistical tests, including likelihood ratio, Engelman-Hartigan, outlier, goodness-of-fit, and normality tests.
  • Considered the entire range of mixing proportions (0 < pi < 1).

Related Experiment Videos

Main Results:

  • For extreme mixing proportions (pi > .85 or pi < .15), Fisher's skewness statistic and Filliben's probability plot correlation coefficient test were most powerful.
  • The Engelman-Hartigan test excelled for moderate mixing proportions (.35 < pi < .65).
  • The likelihood ratio test demonstrated strong performance across other mixing proportions and was consistently powerful (>=50%) when preferred tests achieved similar power.

Conclusions:

  • The optimal statistical test for detecting normal mixture distributions depends on the mixing proportion.
  • The likelihood ratio test offers a robust and generally powerful option across a wide range of mixture scenarios.