Online Program Home
My Program

Abstract Details

Activity Number: 123 - Binary and Ordinal Outcome Regression
Type: Contributed
Date/Time: Monday, July 30, 2018 : 8:30 AM to 10:20 AM
Sponsor: Biometrics Section
Abstract #329706 Presentation
Title: Ordinal Outcomes: Considerations for the Generalized Linear Model with the Log Link
Author(s): Gurbakhshash Singh* and Gordon Hilton Fick
Companies: University of Calgary and University of Calgary
Keywords: Cumulative Probability; Log Link; Proportional Probability; Ordinal Outcomes/Responses; Maximum Likelihood; Uniqueness

There are many options available for the analysis of ordinal outcomes. The Proportional Odds Model, based on the logit link, is, perhaps, still the most often seen. Recently, a comparable model has been suggested based on the log link. Using the log link introduces constraints on the parameter space. Other constraints continue to translate from any link. We present results 1) on conditions for the uniqueness of the Maximum Likelihood Estimate (MLE) based on the rank of certain matrices, 2) about using constrOptim in R to determine the MLE 3) and some other fundamental results. We also offer some closed form expressions for the MLE and we offer some results with hypothesis tests in constrained spaces.

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

Back to the full JSM 2018 program