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

Bayesian latent variable models for mixed discrete outcomes.

David B Dunson1, Amy H Herring

  • 1Biostatistics Branch, National Institute of Environmental Health Sciences, MD A3-03, PO Box 12233, Research Triangle Park, NC 27709, USA. dunson1@niehs.nih.gov

Biostatistics (Oxford, England)
|December 25, 2004
PubMed
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 author

Sequential Gibbs posteriors with applications to principal component analysis.

Biometrika·2026
Same author

Scalable and robust regression models for continuous proportional data.

Journal of the American Statistical Association·2026
Same author

Local graph estimation with pathwise false discovery control.

Nature communications·2026
Same author

Bayesian Transfer Learning.

Statistical science : a review journal of the Institute of Mathematical Statistics·2026
Same author

Domain Adaptive Bootstrap Aggregating.

IEEE transactions on signal processing : a publication of the IEEE Signal Processing Society·2026
Same author

Logistic-Beta Processes for Dependent Random Probabilities with Beta Marginals.

Bayesian analysis·2026
Same journal

Towards optimal environmental policies: policy learning under arbitrary bipartite network interference.

Biostatistics (Oxford, England)·2026
Same journal

Multilevel functional quantile principal component analysis.

Biostatistics (Oxford, England)·2026
Same journal

Adaptive transfer learning for time-to-event modeling with applications in disease risk assessment.

Biostatistics (Oxford, England)·2026
Same journal

High-dimensional test for one-sided hypotheses.

Biostatistics (Oxford, England)·2026
Same journal

NBSR: a Negative Binomial Softmax Regression model for microRNA-seq data analysis.

Biostatistics (Oxford, England)·2026
Same journal

Addressing the influence of unmeasured confounding in observational studies with time-to-event outcomes: a semiparametric sensitivity analysis approach.

Biostatistics (Oxford, England)·2026
See all related articles

This study introduces a Poisson variable framework to analyze mixed discrete health outcomes, like tumor development over time. This flexible model accounts for dependencies, improving the analysis of complex health conditions.

Area of Science:

  • Biostatistics
  • Statistical Modeling
  • Health Research Methodology

Background:

  • Complex health conditions often involve multiple, mixed discrete outcomes (e.g., event time, counts, binary, ordered categorical).
  • Analyzing these mixed outcomes requires sophisticated statistical approaches to account for dependencies.
  • Existing methods may not adequately capture the intricate relationships within mixed discrete data common in health studies.

Purpose of the Study:

  • To propose a unified statistical framework for analyzing mixed discrete outcomes in health research.
  • To develop a model that accommodates dependencies among different types of discrete outcomes.
  • To provide a flexible and robust method for analyzing complex health conditions, exemplified by skin tumorigenesis studies.

Main Methods:

Related Experiment Videos

  • Development of a general underlying Poisson variable framework for mixed discrete outcomes.
  • Incorporation of an additive gamma frailty model to handle dependencies between Poisson means.
  • Application of specific model forms (log-linear, complementary log-log, proportional hazards) for count, binary, and event time data, respectively.
  • Utilizing a Bayesian inference approach with conditionally-conjugate priors and Markov Chain Monte Carlo (MCMC) for posterior computation.

Main Results:

  • The proposed framework successfully models mixed discrete outcomes, including event time, counts, and binary data.
  • The additive gamma frailty model effectively captures dependencies among these outcomes.
  • Closed-form expressions for marginal expectations, variances, and correlations were derived, simplifying analysis.
  • The methodology was illustrated using data from a Tg.AC mouse bioassay study, demonstrating practical applicability.

Conclusions:

  • The proposed Poisson variable framework with gamma frailty offers a unified and flexible approach for analyzing mixed discrete outcomes in health studies.
  • This method enhances the ability to model complex health conditions by accounting for interdependencies.
  • The Bayesian MCMC approach provides an efficient computational strategy for inference.