The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Abstract Details
|
Activity Number:
|
575
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 3, 2011 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #302777 |
|
Title:
|
Linear Latent Structure Analysis (LLS)
|
|
Author(s):
|
Mikhail Kovtun*+ and Igor Akushevich and Anatoliy Yashin
|
|
Companies:
|
Duke University and Duke University and Duke University
|
|
Address:
|
, Durham, NC, 27708,
|
|
Keywords:
|
mixed distributions ;
latent structure ;
identifiability ;
mixture invariants ;
estimation ;
consistency
|
|
Abstract:
|
LLS analysis is aimed to model a joint distribution of a number of categorical random variables (survey-like data) as a mixture of independent distributions. Generally, the mixing distribution cannot be identified. However, if the mixing distribution is carried by a low-dimensional subspace of the space of independent distributions (which is when LLS analysis is applicable), the carrying subspace of the mixing distribution and a number of low-order moments of the mixing distribution can be identified. The number of identifiable moments increases as the number of categorical random variables increases, and in the case of infinite number of random variables the mixing distribution is identifiable. Under modest assumptions, the identifiable invariants of the mixing distribution can be consistently estimated, and there exists an efficient computational algorithm for estimating identifiable i
|
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 2011 program
|
2011 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.