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

Bootstrap tests for overdispersion in a zero-inflated Poisson regression model.

Byoung Cheol Jung1, Myoungshic Jhun, Jae Won Lee

  • 1Department of Statistics, Sungshin University, Seoul 136-742, Korea.

Biometrics
|July 14, 2005
PubMed
Summary

A new parametric bootstrap method improves hypothesis testing for zero-inflated models. This approach maintains accurate significance levels and offers greater statistical power compared to normal approximations, especially in small samples.

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

Enhancement of Collagen and Elastic Fibers by Direct Contact Cooling-Based Monopolar Radiofrequency in a Minipig Model.

Clinical, cosmetic and investigational dermatology·2026
Same author

Nonparametric testing methods based on relative effect in non-inferiority clinical trial with multiple experimental drugs.

Journal of biopharmaceutical statistics·2026
Same author

Pulsed-dye laser therapy for successful management of refractory neurogenic rosacea.

Journal of cosmetic and laser therapy : official publication of the European Society for Laser Dermatology·2026
Same author

Active nitrogen mediated selective ruthenium migration on ceria for high pressure ammonia decomposition.

Nature communications·2026
Same author

Machine learning reveals microbiome differences by periodontitis severity.

PloS one·2026
Same author

Controlled Formation of Polyimide Aerogel Networks in Carbon Fiber Felt via Multicycle Freeze-Drying for Thermal Protection.

Polymers·2026

Area of Science:

  • Biostatistics
  • Statistical modeling
  • Count data analysis

Background:

  • Zero-inflated regression models are crucial for analyzing count data with excess zeros.
  • Existing score tests for zero-inflated Poisson (ZIP) versus zero-inflated negative binomial (ZINB) models may lack accuracy in small samples.
  • Normal approximation in score tests can lead to underestimated significance levels.

Discussion:

  • The proposed parametric bootstrap method addresses the limitations of normal approximation for score tests.
  • This bootstrap approach ensures the significance level remains close to the nominal level.
  • It demonstrates uniformly greater power than the normal approximation method.

Key Insights:

  • Parametric bootstrap offers a robust solution for hypothesis testing in zero-inflated models.

Related Experiment Videos

  • Improved accuracy and power in significance testing for ZIP vs. ZINB models.
  • Essential for reliable statistical inference with small sample sizes in biostatistics.

    • Outlook:

    • Potential for broader application in other complex statistical models with excess zeros.
    • Further research into optimizing bootstrap methods for various count data distributions.
    • Enhancing the reliability of statistical tests in epidemiological and biological studies.