|
Activity Number:
|
61
- Approaches for Modeling Clustered and Longitudinal Data
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Biometrics Section
|
|
Abstract #312253
|
|
|
Title:
|
Conway-Maxwell-Multinomial Regression for Categorical Data with Associated Trials
|
|
Author(s):
|
Darcy Morris* and Andrew M. Raim and Kimberly Sellers
|
|
Companies:
|
and U.S. Census Bureau and Georgetown University
|
|
Keywords:
|
Multinomial Regression;
COM-Poisson;
Count Data;
Clustered Data;
Categorical Data
|
|
Abstract:
|
Categorical data are often observed as counts resulting from a fixed number of trials in which each trial consists of making one selection from a prespecified set of categories. Multinomial regression serves as a standard model for such clustered data but assumes that trials are independent and identically distributed given the covariates. This work considers Conway-Maxwell-multinomial (CMM) regression for modeling clustered categorical data exhibiting positively or negatively associated trials. The CMM distribution features a dispersion parameter which allows it to adapt to a range of association levels that may depend on observed explanatory variables. Using public data from the U.S. Census Bureau, we describe a CMM regression model of a household categorical variable that may exhibit geographic association. The results illustrate insight gained from estimating characteristics of trial-level association.
|
Authors who are presenting talks have a * after their name.
