|
Activity Number:
|
106
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
ENAR
|
| Abstract - #309108 |
|
Title:
|
Principal Component Analysis for Multivariate Binary Data
|
|
Author(s):
|
Seokho Lee*+
|
|
Companies:
|
Texas A&M University
|
|
Address:
|
3143 TAMU, College Station, TX, 77843-3143,
|
|
Keywords:
|
|
|
Abstract:
|
Principal component analysis (PCA) is frequently used as a powerful tool for dimension reduction in multivariate data analysis. One of its disadvantages is that PCA is applicable only to continuous data. There is little literature on PCA-like dimension reduction methods for multivariate binary data which are commonly observed in various application fields. The nature of binary data circumvents a direct application of PCA to multivariate binary data. In this presentation, we propose an extension of PCA to handle multivariate binary variables by using Bernoulli distribution and low-rank approximations of matrices of latent variables. Illustrations of application to real and synthetic datasets are presented.
|
- 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 2007 program |
