|
Activity Number:
|
318
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Mining
|
| Abstract - #305494 |
|
Title:
|
Penalized Rotation of a Subset of Principal Components
|
|
Author(s):
|
Trevor Park*+
|
|
Companies:
|
University of Florida
|
|
Address:
|
Department of Statistics, Gainesville, FL, 32611,
|
|
Keywords:
|
penalized likelihood ; PCA
|
|
Abstract:
|
Traditional principal component rotation restricts the rotation to the span of a select subset of the original components. Since this subspace is subject to sampling variability, it may not capture anticipated features of the true components, and the rotation will be less effective. Penalty-based methods remove this restriction. A Bayesian-motivated penalized method will be described, along with computational details. This method preserves orthogonality, unlike many competing methods. Similar methodology can be applied to rotation in traditional factor analysis.
|
- 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 2009 program |
