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

Parametric bootstrap and penalized quasi-likelihood inference in conditional autoregressive models.

Y C MacNab1, C B Dean

  • 1Centre for Health Evaluation Research, British Columbia Institute for Children's and Women's Health, 4480 Oak Street, Rm E-414, Vancouver, B.C., Canada V6H 3V4. ymacnab@cw.bc.ca

Statistics in Medicine
|August 29, 2000
PubMed
Summary

This study introduces advanced conditional autoregressive (CAR) models for disease mapping, offering a robust algorithm for health agencies. The methods provide reliable inference for spatial autocorrelation, improving disease rate analysis.

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

A discrete-time susceptible-infectious-recovered-susceptible model for the analysis of influenza data.

Infectious Disease Modelling·2023
Same author

Spatial statistical tools for genome-wide mutation cluster detection under a microarray probe sampling system.

PloS one·2018
Same author

Classification of Large-Scale Remote Sensing Images for Automatic Identification of Health Hazards: Smoke Detection Using an Autologistic Regression Classifier.

Statistics in biosciences·2017
Same author

Joint modeling of zero-inflated panel count and severity outcomes.

Biometrics·2017
Same author

Efficient panel designs for longitudinal recurrent event studies recording panel counts.

Biostatistics (Oxford, England)·2013
Same author

The consequences of proportional hazards based model selection.

Statistics in medicine·2013

Area of Science:

  • Spatial statistics
  • Biostatistics
  • Geographic information systems (GIS)

Background:

  • Traditional conditional autoregressive (CAR) models are widely used for disease mapping.
  • Existing models may not fully capture complex data structures in disease rate analysis.
  • Accurate spatial modeling is crucial for public health surveillance and resource allocation.

Purpose of the Study:

  • To extend conditional autoregressive (CAR) models beyond the standard first-order intrinsic CAR model.
  • To present a practical algorithm for fitting advanced CAR models for disease rate mapping.
  • To enhance the reliability of statistical inference for spatial autocorrelation in health data.

Main Methods:

  • Utilized penalized quasi-likelihood (PQL) inference for parameter estimation.

Related Experiment Videos

  • Employed an analogue of best-linear unbiased estimation for regional risk ratios.
  • Applied restricted maximum likelihood for variance component estimation.
  • Introduced the parametric bootstrap for robust statistical inference, especially for boundary hypotheses.

    • Main Results:

    • Demonstrated the utility of advanced CAR models for diverse data structures.
    • Presented a straightforward algorithm for routine use in statistical and health agencies.
    • Showcased the parametric bootstrap's reliability over standard maximum likelihood asymptotics.
    • Illustrated the methodology with an infant mortality analysis in British Columbia, Canada.

    Conclusions:

    • Advanced CAR models offer enhanced capabilities for disease rate mapping.
    • The proposed PQL-based algorithm and parametric bootstrap provide reliable tools for practitioners.
    • The methodology effectively handles spatial autocorrelation, improving the accuracy of health maps.
    • This work facilitates routine use of sophisticated spatial models in public health.