|
Activity Number:
|
193
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
WNAR
|
| Abstract - #309583 |
|
Title:
|
M-Type Smoothing Spline ANOVA for Correlated Data
|
|
Author(s):
|
Anna Liu*+ and John Staudenmayer and Li Qin
|
|
Companies:
|
University of Massachusetts Amherst and University of Massachusetts Amherst and Fred Hutchinson Cancer Research Center
|
|
Address:
|
, Amherst, MA, 01003,
|
|
Keywords:
|
Robust ; Smoothing spline ANOVA ; Correlated data ; Resistant smoothing parameter ; Robust inference ; Hypothesis test
|
|
Abstract:
|
This paper extends the M-type smoothing spline regression models by Huber (1979) for independent observations with a single smoothing parameter to smoothing spline ANOVA models for correlated data. Simultaneous resistant estimates of the parameterized covariance matrix and the smoothing parameters are developed. This framework allows robust inference on the regression function. Inference procedures are developed for general hypotheses and evaluated through simulations for testing a constant regression function and no difference of mean group curves. Simulations show that the robust tests have improved performances over their Gaussian counterparts. Applications to real data will be demonstrated.
|
- 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 |
