Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 521 - Advances in Methods and Novel Applications for Data with Hierarchies and Under/Over-Dispersion
Type: Invited
Date/Time: Thursday, August 6, 2020 : 1:00 PM to 2:50 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #309229
Title: Analyzing Continuous Longitudinal Response Data with Ordinal Regression Models
Author(s): Bryan E Shepherd* and Yuqi Tian and Jonathan Schildcrout
Companies: Vanderbilt University and Vanderbilt University and Vanderbilt University
Keywords: ordinal; longitudinal data; semiparametric models

Continuous data can be analyzed with ordinal cumulative probability models (CPMs), also known as 'cumulative link models.' These models belong to the class of semiparametric linear transformation models, where the data are assumed to follow a known linear model after some unspecified transformation that is empirically estimated. CPMs have many benefits including that they are able to handle a wide variety of data types; they can readily address detection limits; they are invariant to monotonic transformations of the response variable; and they model the cumulative distribution function from which conditional expectations, quantiles, probability indices, and other interpretable parameters are easily derived. To date, most research with these models has focused on cross-sectional data. We investigate the use of CPMs with repeated measures longitudinal data. We describe extensions of these models using generalized estimation equations approaches, and we discuss computational challenges.

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

Back to the full JSM 2020 program