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

Approximate cross-validatory predictive checks in disease mapping models.

E C Marshall1, D J Spiegelhalter

  • 1Department of Epidemiology and Public Health, Imperial College of Science, Technology and Medicine, Norfolk Place, London W2 1PG, UK. c.marshall@ic.ac.uk

Statistics in Medicine
|April 30, 2003
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

Statistics. The future lies in uncertainty.

Science (New York, N.Y.)·2014
Same author

The power of the MicroMort.

BJOG : an international journal of obstetrics and gynaecology·2014
Same author

Using routine intelligence to target inspection of healthcare providers in England.

Quality & safety in health care·2009
Same author

How to interpret your dot: decoding the message of clinical performance indicators.

Journal of perinatology : official journal of the California Perinatal Association·2008
Same author

Flexible random-effects models using Bayesian semi-parametric models: applications to institutional comparisons.

Statistics in medicine·2006
Same author

Intracortically distributed neurovascular coupling relationships within and between human somatosensory cortices.

Cerebral cortex (New York, N.Y. : 1991)·2006

This study improves model assessment for complex disease mapping. Replicating random effects and data offers a faster, more effective way to identify regions deviating from Bayesian hierarchical models.

Area of Science:

  • Biostatistics
  • Spatial Epidemiology
  • Bayesian Modeling

Background:

  • Complex hierarchical disease mapping models are crucial for epidemiological studies.
  • Identifying regions that deviate from model assumptions is vital for accurate disease mapping.
  • Traditional leave-one-out cross-validation is computationally intensive with Markov chain Monte Carlo (MCMC) methods.

Purpose of the Study:

  • To improve upon existing importance sampling approximations for assessing complex hierarchical disease mapping models.
  • To develop a more efficient method for identifying regions that diverge from model assumptions.
  • To provide a generic and easily applicable approach for criticizing Bayesian hierarchical models.

Main Methods:

  • The study builds upon the importance sampling approximation proposed by Stern and Cressie.

Related Experiment Videos

  • The proposed method involves replication of both random effects and data within the model fitting process.
  • This approach is designed to be compatible with Markov chain Monte Carlo (MCMC) estimation.

    • Main Results:

    • The proposed method of replicating random effects and data improves upon the standard importance sampling approximation.
    • This enhanced approach provides a more computationally feasible way to perform cross-validatory assessments.
    • The method is shown to be effective in identifying regions that do not fit the assumed hierarchical model.

    Conclusions:

    • Replication of random effects and data offers a practical enhancement to model assessment in Bayesian hierarchical disease mapping.
    • This generic and simple-to-apply method can aid in the critical evaluation of complex spatial epidemiological models.
    • The findings contribute to more robust and reliable disease mapping analyses.