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

A score test for zero inflation in a Poisson distribution

J van den Broek1

  • 1Center for Biostastics, University of Utrecht, The Netherlands.

Biometrics
|June 1, 1995
PubMed
Summary

Excessive zeros in count data may indicate a need for a zero-inflated Poisson distribution. This study presents a score test to determine if zero counts are too high for a standard Poisson model.

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

Periparturient oleic acid-rich fat supplementation affects the lipid profile in blood and results in an increased oocyte yield in postpartum dairy cows.

Theriogenology·2025
Same author

A model-based approach to evaluate the effect of vaccination of the herd on transmission of equine herpesvirus 1 in naturally occurring outbreaks.

Preventive veterinary medicine·2025
Same author

Heat stress in a temperate climate leads to adapted sensor-based behavioral patterns of dairy cows.

Journal of dairy science·2022
Same author

Highly selective detection of methanol over ethanol by a handheld gas sensor.

Nature communications·2019
Same author

Sensor based eating time variables of dairy cows in the transition period related to the time to first service.

Preventive veterinary medicine·2019
Same author

Factors affecting the likelihood of pregnancy and embryonic loss after transfer of cryopreserved in vitro produced equine embryos.

Equine veterinary journal·2018

Area of Science:

  • Statistics
  • Biostatistics
  • Data Analysis

Background:

  • Count data frequently exhibit an excess of zero observations.
  • Standard Poisson distribution may not adequately model datasets with numerous zeros.
  • Zero-inflated models offer an alternative for such data.

Purpose of the Study:

  • To introduce a score test for assessing excess zeros in count data.
  • To evaluate the suitability of a Poisson distribution versus a zero-inflated Poisson distribution.
  • To provide a statistical method for identifying data requiring zero-inflated modeling.

Main Methods:

  • Development of a score test statistic.
  • Application of the test to count data with many zeros.
  • Comparison of model fit between Poisson and zero-inflated Poisson distributions.

Main Results:

  • The score test effectively detects situations where zero counts are excessively high for a Poisson model.
  • Identifies datasets that would benefit from the application of zero-inflated Poisson regression.
  • Provides a quantitative measure to guide model selection.

Conclusions:

  • A score test is a valuable tool for analyzing count data with excess zeros.
  • Facilitates appropriate statistical modeling by distinguishing between Poisson and zero-inflated Poisson data.
  • Aids researchers in selecting the most accurate distribution for their specific count data challenges.

Related Experiment Videos