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

Bivariate random effect model using skew-normal distribution with application to HIV-RNA.

Pulak Ghosh1, Marcia D Branco, Hrishikesh Chakraborty

  • 1Department of Mathematics and Statistics, Georgia State University, Atlanta, GA 30303-3083, USA. pghosh@mathstat.gsu.edu

Statistics in Medicine
|September 26, 2006
PubMed
Summary

This study introduces a Bayesian bivariate mixed model using a multivariate skew-normal distribution to better analyze correlated data in longitudinal studies when normality assumptions fail. The approach offers more robust results for skewed data, as demonstrated with an HIV study dataset.

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

Hydrocodone vs Oxycodone and Postoperative Pain and Opioid Use in Joint Arthroplasty.

JAMA network open·2026
Same author

CYP2D6 phenotype and post-surgical pain control with hydrocodone and oxycodone.

Frontiers in pharmacology·2026
Same author

Genotype-Guided Antidepressant Prescribing for Patients With Depression: A Randomized Clinical Trial.

JAMA network open·2026
Same author

Genetic Testing for APOL1 in Adults With Hypertension: The GUARDD-US Randomized Clinical Trial.

JAMA network open·2026
Same author

Impact of COVID 19 pandemic on young patients with acute coronary syndrome in Eastern Indian population.

Indian heart journal·2026
Same author

Perspectives from Sri Lankan physicians on current approaches to diagnosing and managing lower respiratory tract infections.

BMC research notes·2026

Area of Science:

  • Biostatistics
  • Epidemiology
  • Clinical Research

Background:

  • Correlated data are common in longitudinal studies, particularly in epidemiological and clinical research.
  • Random effects models are standard for analyzing correlated data, often assuming normal distributions for random effects and errors.
  • Normality assumptions can lead to non-robust results when longitudinal data exhibit skewness.

Purpose of the Study:

  • To develop a Bayesian approach for bivariate mixed models that relaxes the normality assumption.
  • To incorporate a multivariate skew-normal distribution to handle skewed data in longitudinal settings.
  • To compare different modeling strategies and demonstrate the proposed method using real-world HIV study data.

Main Methods:

  • Development of a Bayesian bivariate mixed model.

Related Experiment Videos

  • Application of a multivariate skew-normal distribution to relax normality assumptions.
  • Comparative analysis of various potential models.
  • Illustration using a real dataset from an HIV study.

    • Main Results:

    • The proposed Bayesian approach provides a flexible alternative to standard normal-based random effects models.
    • The multivariate skew-normal distribution effectively accommodates skewness in correlated longitudinal data.
    • The model comparison and application demonstrate the practical utility of the method.

    Conclusions:

    • The Bayesian bivariate mixed model with a multivariate skew-normal distribution offers a robust framework for analyzing skewed correlated data in longitudinal studies.
    • This approach enhances the reliability of analyses in epidemiological and clinical research where normality may not hold.
    • The methodology is validated through its application to a relevant HIV study dataset.