Order restricted inferences on parameters in generalized linear models with emphasis on logistic regression

Resource Information

The work Order restricted inferences on parameters in generalized linear models with emphasis on logistic regression represents a distinct intellectual or artistic creation found in University of Missouri Libraries. This resource is a combination of several types including: Work, Language Material, http://bibfra.me/vocab/marc/Manuscript, Books.

Label

Order restricted inferences on parameters in generalized linear models with emphasis on logistic regression

Statement of responsibility

by Diann Reischman

Title variation

Order restricted inferences in GLMS

Creator

• Reischman, Diann, (Diann Elizabeth), 1963-

Subject

• Microfilm.
• Dissertations, Academic -- University of Missouri--Columbia -- Statistics
• Electronic books
• Regression analysis
• Mathematical statistics
• Probabilities
• Order statistics
• Parameter estimation
• Linear models (Statistics)

Language

eng

Summary

In a generalized linear model, a likelihood ratio test is generally performed to test for significance of all or part of the parameter set. If there are some order restrictions on any of the parameters in the portion being tested, a test that uses this information may be more powerful than the usual likelihood ratio test. We concentrate on generalized linear models with a Bernoulli response variable. In particular, testing for significance of a set of parameters using the order restrictions. Monte Carlo simulations are performed on the one-variable and two-variable logistic regression models to investigate these five tests in this context. Two real-life examples are shown to illustrate these five tests. Simulations are also performed on one-variable models using the probit link and the complementary log-log link to investigate the behavior of the tests when a link other than the logit is used. Extensions to a generalized linear model with a Poisson response variable are mentioned

Additional physical form

Also available on the Internet.

Cataloging source

MUU

Degree

Ph. D.

Dissertation year

1997.

Government publication

government publication of a state province territory dependency etc

Granting institution

University of Missouri-Columbia

Illustrations

illustrations

Index

no index present

Literary form

non fiction

Nature of contents

bibliography

Target audience

specialized