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

Conditional independence test by generalized Kendall's tau with generalized odds ratio.

Shuang Ji1, Jing Ning2, Jing Qin3

  • 11 Morgan Stanle, New York, NY, USA.

Statistical Methods in Medical Research
|January 5, 2018
PubMed
Summary

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

Searching for immune correlates in Lassa vaccine development - workshop report.

NPJ vaccines·2026
Same author

Correlates of severe and delta COVID-19 in a phase 3 trial of the AZD1222 vaccine.

NPJ vaccines·2026
Same author

Continuous Infusion of the CXCR4 Antagonist Plerixafor for WHIM Syndrome.

Journal of clinical immunology·2026
Same author

Impact of prior SARS-CoV-2 acquisition on binding and neutralizing antibody responses following COVID-19 vaccination: A cross-protocol analysis of individual-level data from six phase 3 clinical trials.

Vaccine·2026
Same author

Plasma biomarkers of immunothrombosis are independently associated with death in patients with COVID-19 on ECMO.

The Journal of infectious diseases·2026
Same author

Association between COVID-19 vaccine efficacy and epidemic force of infection.

NPJ vaccines·2026

This study introduces a new method for testing conditional independence, overcoming the curse of dimensionality without restricting variable distributions. The generalized weighted Kendall's tau offers a powerful tool for complex data analysis.

Area of Science:

  • Statistics
  • Machine Learning
  • Bioinformatics

Background:

  • Determining conditional dependence is crucial for model building and applications like genetic association studies.
  • Existing methods often rely on kernel-based approaches or categorical variables, facing challenges with high-dimensional data (curse of dimensionality).

Purpose of the Study:

  • To propose a novel class of tests for conditional independence that do not require restrictions on the distribution of conditioning variables.
  • To address the limitations of existing methods in handling high-dimensional and continuous conditioning variables.

Main Methods:

  • Developed a test statistic based on a generalized weighted Kendall's tau.
  • Utilized a generalized odds ratio as a weight function to incorporate distances between conditioning variable values.
Keywords:
Conditional independenceU-statisticsgeneralized Kendall’s taugeneralized odds ratio

Related Experiment Videos

  • The proposed test procedure exhibits desirable asymptotic properties and is straightforward to implement.

    • Main Results:

    • The proposed test effectively determines conditional independence without distributional assumptions on conditioning variables.
    • Simulation studies demonstrated favorable finite sample performance.
    • The method was successfully illustrated using two real-world data examples.

    Conclusions:

    • The novel test offers a flexible and implementable solution for conditional independence testing, particularly in high-dimensional settings.
    • This approach advances statistical modeling and applications in fields like graphical models and genetic association studies.