|
Activity Number:
|
17
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
IMS
|
| Abstract - #308136 |
|
Title:
|
Generalized K-Means Inverse Regression Estimation
|
|
Author(s):
|
Xuerong Wen*+
|
|
Companies:
|
University of Missouri-Rolla
|
|
Address:
|
220 Rolla Bldg, Rolla, MO, 65409,
|
|
Keywords:
|
Dimension Reduction ; Multivariate Regression ; Central Subspace ; K-means Clustering ; Intra-cluster Information ; IRE
|
|
Abstract:
|
Few methodologies are available for estimating the central subspace for regressions with multivariate responses due to the difficulties arising from slicing multivariate responses. Setodji and Cook(2004) introduced a new way of performing the slicing. They developed a method called k-means inverse regression (KIR), which makes use of the K-means algorithm to cluster the observed response vectors. However, their method ignored the intra-cluster information which could be substantial under some circumstances. In this paper, we proposed an improved method by incorporating the intra-cluster information into estimation. Our method outperformed KIR with respect to estimation accuracies of both the central subspace and its dimension. It also allows us to test the predictor effects in a model-free approach.
|
- 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 |
