|
Activity Number:
|
157
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #304633 |
|
Title:
|
Penalized Regression, Mixed Effects Models, and Appropriate Modeling
|
|
Author(s):
|
Nancy E. Heckman*+ and Richard Lockhart and Jason D. Nielsen
|
|
Companies:
|
The University of British Columbia and Simon Fraser University and Carleton University
|
|
Address:
|
Statistics Department, Vancouver, BC, V6T1Z2, Canada
|
|
Keywords:
|
linear mixed effects model ; penalized smoothing ; P-spline ; sandwich estimator
|
|
Abstract:
|
Linear mixed effects methods for the analysis of longitudinal data provide a convenient framework for modeling within-individual correlation across time. Using spline functions allows for flexible modeling of the response as a function of time. A computational connection between linear mixed effects modeling and spline smoothing makes the use of spline functions in longitudinal data analysis even more appealing. However, care must be taken in exploiting this connection, as resulting estimates of the underlying population mean might not track the data well and associated standard errors might be unreasonably large. We discuss these shortcomings and suggest some easy-to-compute methods to eliminate them.
|
