Univariate polytomous ordinal regression analysis with application to diabetic retinopathy data

Batten, Dennis William (2000) Univariate polytomous ordinal regression analysis with application to diabetic retinopathy data. Masters thesis, Memorial University of Newfoundland.

[img] [English] PDF (Migrated (PDF/A Conversion) from original format: (application/pdf)) - Accepted Version
Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

Download (8MB)
  • [img] [English] PDF - Accepted Version
    Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.
    (Original Version)

Abstract

There are many situations in practice where one is interested to find the regression effects of the covariates on polytontous responses. Furthermore, there are situations where polytomous responses are ordinal by nature. These types of data are commonly analyzed by exploiting the well-known probit and cumulative logit models. These methods, however, require the introduction of certain cut-points to distinguish ordered categories of the polytomous responses, and these cut-points are required ito be estimated consistently, which may not be easily obtained. In the practncum, we use a recently developed non-cut-point based cumulative logit mo del to resolve this estimation problem. The regression analysis chosen in the practicum was motivated by a need for a refined analysis of a diabetes data set used in the Wisconsin Epidemiologic Study of Diabetic Retinopatlhy (WESDR). The practicum discusses the advantages and disadvantages of the existing as well as the new techniques- The non-cut-point based approach was found to give the best fit to the diabetes data, with easy interpretation of the regression estimates.

Item Type: Thesis (Masters)
URI: http://research.library.mun.ca/id/eprint/747
Item ID: 747
Additional Information: Bibliography: leaves 70-71
Department(s): Science, Faculty of > Mathematics and Statistics
Date: 2000
Date Type: Submission
Library of Congress Subject Heading: Regression analysis; Probits; Diabetic retinopathy

Actions (login required)

View Item View Item

Downloads

Downloads per month over the past year

View more statistics