JSM 2015 Preliminary Program

Online Program Home
My Program

Abstract Details

Activity Number: 317
Type: Contributed
Date/Time: Tuesday, August 11, 2015 : 8:30 AM to 10:20 AM
Sponsor: SSC
Abstract #317015 View Presentation
Title: New Powerful Goodness-of-Fit Tests for Ordinal Logistic Regressions
Author(s): Zheng Sun*
Companies: Simon Fraser University
Keywords: ordinal logistic regression ; goodness-of-fit ; ordinal categorical data ; likelihood ratio tests ; EDF tests
Abstract:

Generalized linear models are widely used to model data with discrete responses and the standard goodness-of-tests for these ordinal models are the chi-square test and the deviance test. When the response has multiple categories that have obvious natural ordering, McCullagh (1980) suggested modeling the logit of the cumulative probabilities instead of the probabilities. This takes into account the ordering of the responses. But for testing, the chi-square test and the deviance test are not order sensitive. In this paper, new tests are developed that are based on the cumulative probabilities and therefore are order sensitive. Asymptotic distributions of the test statistics are given and can be used to conduct the tests. Power studies show the new tests give better power than the standard goodness-of-fit statistics against a wide range of alternatives.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2015 program





For program information, contact the JSM Registration Department or phone (888) 231-3473.

For Professional Development information, contact the Education Department.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

2015 JSM Online Program Home