Search research articles

Search research articles

Related Experiment Videos

Mixture models, robustness, and the weighted likelihood methodology.

¹Department of Statistics, Columbia University, New York, New York 10027, USA. markat@stat.columbia.edu

|July 6, 2000

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 journal

Interim analysis in sequential multiple assignment randomized trials for survival outcomes.

Biometrics·2026

Same journal

Acknowledgment of Referees 2025.

Biometrics·2026

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

See all related articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Weighted likelihood estimation effectively handles mixture model analysis challenges like outliers and misspecification. This statistical method offers low bias and mean squared error, revealing data substructures for improved model fitting.

Area of Science:

Statistics
Data Analysis
Machine Learning

Background:

Analyzing data from mixture distributions presents challenges including outliers, underrepresented components, and model misspecification biases.
Existing methods may struggle with these complexities, leading to inaccurate parameter estimation and model interpretation.

Purpose of the Study:

To evaluate the performance of weighted likelihood estimation for analyzing mixture distribution data.
To address practical issues in implementing weighted likelihood, such as starting value selection and bootstrap sample size.
To introduce a novel statistical stopping rule for algorithm termination.

Main Methods:

Weighted likelihood estimation is employed to address problems in mixture model analysis.
Method of moment estimates from bootstrap subsamples are used as starting values for weighted likelihood computation.

Related Experiment Videos

A new statistical stopping rule is proposed for algorithm convergence.

Main Results:

Weighted likelihood estimation yields estimates with low bias and mean squared error.
The method successfully unearths data substructures, indicated by multiple roots, suggesting potential for more components than initially specified.
The weighted likelihood algorithm demonstrates competitive performance against the Expectation-Maximization (EM) algorithm, particularly when components are not well separated.

Conclusions:

Weighted likelihood estimation is a robust technique for mixture model analysis, effectively mitigating issues like outliers and misspecification.
The method's ability to detect underlying data substructures enhances the identification of appropriate mixture models.
The proposed statistical stopping rule offers a reliable criterion for algorithm termination.