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

Sensitivity of score tests for zero-inflation in count data.

Andy H Lee1, Liming Xiang, Wing K Fung

  • 1Department of Epidemiology and Biostatistics, School of Public Health, Curtin University of Technology, Perth, Australia. Andy.Lee@curtin.edu.au

Statistics in Medicine
|August 19, 2004
PubMed
Summary

Zero-inflated models are useful for biomedical count data with excess zeros. This study introduces diagnostic measures to assess how individual data points influence zero-inflation tests, ensuring reliable model selection.

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

Learning association from multiple intermediate events for dynamic prediction of survival: an application to cardiovascular disease prognosis.

Biometrics·2026
Same author

A double-semiparametric approach for extending mixture cure models with interval-censored data.

Statistical methods in medical research·2026
Same author

Regression analysis of interval-censored competing risks data with missing causes of failure: A direct likelihood approach.

Statistical methods in medical research·2026
Same author

Testing menstrual blood for human papillomavirus during cervical cancer screening in China: cross sectional population based study.

BMJ (Clinical research ed.)·2026
Same author

Formal Remote Stereodifferentiation of Minimally Distinct Groups in the Enantioselective Desymmetrization of <i>meso</i>-Epoxides.

The Journal of organic chemistry·2026
Same author

Dermatomyositis-Like Lesion: An Atypical Cutaneous Presentation in Adult-Onset Still Disease.

Dermatology practical & conceptual·2025

Area of Science:

  • Biostatistics
  • Statistical Modeling

Background:

  • Count data in biomedical applications often feature a high proportion of zeros.
  • Zero-inflated Poisson (ZIP) and zero-inflated negative binomial models are used for such data.
  • Existing score tests for zero-inflation can be sensitive to outliers.

Purpose of the Study:

  • To develop diagnostic measures for assessing the influence of observations on zero-inflation score tests.
  • To improve the reliability of model selection between standard and zero-inflated count models.

Main Methods:

  • Derivation of diagnostic measures to evaluate observation influence on score statistics.
  • Application of these diagnostics to real-world biomedical count data examples.
  • Sensitivity analysis of zero-inflation tests in the context of regression models.

Related Experiment Videos

Main Results:

  • Diagnostic measures were developed to identify influential observations.
  • Demonstrated how outliers can impact the results of zero-inflation tests.
  • Highlighted the importance of sensitivity analysis for accurate model choice.

Conclusions:

  • The proposed diagnostic measures are crucial for robust inference in zero-inflated count data analysis.
  • Sensitivity analysis enhances the reliability of choosing between Poisson, ZIP, and zero-inflated negative binomial models.
  • Ensures accurate biomedical application of count data models by addressing outlier effects.