|
Activity Number:
|
68
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #304315 |
|
Title:
|
Observed Confidence Levels for Principal Component Analysis
|
|
Author(s):
|
Alan M. Polansky*+
|
|
Companies:
|
Northern Illinois University
|
|
Address:
|
Division of Statistics, De Kalb, IL, 60115,
|
|
Keywords:
|
bootstrap ; normal model ; nonparametric ; multivariate ; confidence levels ; confidence measures
|
|
Abstract:
|
In multivariate analysis, the method of principal components is often used as a dimension reduction technique by identifying a set of orthogonal vectors along which the majority of the variation of a multivariate data set lies. Of importance in such an analysis is the number of directions that contain a specified proportion of the total variation, and the directions of these vectors. This talk will focus on applying the method of observed confidence levels to problems involving principal components. In particular, we will focus on assigning levels of confidence to the number of components that explain a specified proportion of variation in a principal components analysis. The talk will include example applications.
|
