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 Monte Carlo conditional inference in exponential families.

J E Kolassa1, M A Tanner

  • 1Department of Biostatistics, University of Rochester, New York 14642, USA. kolassa@biol.bst.rochester.edu

Biometrics
|April 25, 2001
PubMed
Summary

This study introduces a novel algorithm for approximate frequentist conditional inference in Generalized Linear Models (GLIMs). The method offers accurate statistical inference for multiple parameters without Markov chain simulation.

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

Cardiac T2* magnetic resonance for prediction of cardiac complications in thalassemia major.

Circulation·2009
Same author

[Hospitalization period and nutritional status in hospitalized patients].

Nutricion hospitalaria·2007
Same author

A randomized, placebo-controlled, double-blind trial of the effect of combined therapy with deferoxamine and deferiprone on myocardial iron in thalassemia major using cardiovascular magnetic resonance.

Circulation·2007
Same author

Myocardial infarction following sickle cell chest syndrome.

British journal of haematology·2006
Same author

Myocardial iron loading in patients with thalassemia major on deferoxamine chelation.

Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance·2006
Same author

Percutaneous technique for the reduction of knotted coronary catheters.

Heart (British Cardiac Society)·2003

Area of Science:

  • Statistics
  • Statistical Inference
  • Computational Statistics

Background:

  • Accurate statistical inference is crucial for regression models.
  • Existing methods for conditional inference in Generalized Linear Models (GLIMs) are limited.
  • Commercial software often restricts accurate inference to specific cases like logistic regression.

Purpose of the Study:

  • To develop an algorithm for approximate frequentist conditional inference on multiple parameters within any GLIM family.
  • To extend accurate inference capabilities beyond current software limitations.
  • To provide an alternative to existing Markov chain-based methods.

Main Methods:

  • Utilizes double saddlepoint approximations for conditional cumulative distribution functions.

Related Experiment Videos

  • Employs noniterative Monte Carlo methods to approximate joint distributions.
  • Generates samples from an approximate joint distribution of sufficient statistics.

    • Main Results:

    • The algorithm provides accurate approximate frequentist conditional inference for GLIMs.
    • It extends accurate inference to a broader range of regression models.
    • The method avoids the need for Markov chain construction and convergence checks.

    Conclusions:

    • The proposed algorithm offers a computationally efficient and accurate approach to conditional inference in GLIMs.
    • It presents a viable alternative to existing simulation-based methods.
    • Demonstrated applicability to logistic and truncated Poisson regression models.