|
Activity Number:
|
366
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
IMS
|
| Abstract - #305660 |
|
Title:
|
Sparse Principal Component Analysis
|
|
Author(s):
|
Hui Zou*+
|
|
Companies:
|
University of Minnesota
|
|
Address:
|
313 Ford Hall, Minneapolis, MN, 55455,
|
|
Keywords:
|
PCA ; SPCA ; regularization ; L_1 penalty ; Procrustes rotation ; LARS
|
|
Abstract:
|
Principal component analysis is used widely in data processing and dimensionality reduction. However, PCA suffers from each principal component being a linear combination of the original variables. Thus, it is often difficult to interpret the results. We introduce the SPCA criterion, which results in a principled method---called SPCA---for producing modified principal components with sparse loadings. We also propose an efficient algorithm based on reduced-rank Procrustes rotation and the LARS for computing the SPCA. The proposed methodology is applied to real and simulated data with encouraging results.
|
- 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 2006 program |
