Activity Number:
|
183
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 13, 2002 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistical Computing*
|
Abstract - #301859 |
Title:
|
Estimating Random Effects in Multidimensional Unfolding
|
Author(s):
|
Shiang-Tung Jung*+ and Richard Gonzalez
|
Affiliation(s):
|
University of Michigan and University of Michigan
|
Address:
|
, , , ,
|
Keywords:
|
multidimensional scaling ; random effect ; multidimensional unfolding ; alternative least squares
|
Abstract:
|
Multidimensional unfolding is a scaling technique for rating data. In multidimensional unfolding, each subject gives ratings for all stimuli. Greenacre and Browne (1986) proposed an alternating least-squares algorithm to fit a metric model to estimate subject ideal points and configuration of stimuli. Because subjects are randomly selected, the subject ideal points should be treated as random effects. We propose a modification of the Greenacre and Browne algorithm that allows random effects parameters. The benefits of this model are the ability to estimate random effect within subjects and the ability to include linear predictors.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2002 program |