The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
658
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 2, 2012 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Biometrics Section
|
Abstract - #306335 |
Title:
|
Conditional-Mean Model Selection with Multiple Covariates and Missing Data
|
Author(s):
|
Gregory DiRienzo*+
|
Companies:
|
|
Address:
|
University at Albany - SUNY, Rensselaer, NY, 12144, United States
|
Keywords:
|
Bootstrap ;
Cross-Validation ;
Generalized linear model ;
Mis-specified model ;
Multiple hypothesis testing ;
Multiple hypothesis testing
|
Abstract:
|
An objective methodology to select a parsimonious conditional-mean model from multiple covariates and missing data is proposed. The methods assume data is missing at random and employ inverse probability weighted complete case estimation of regression coefficients and cross-validated prediction error for working models. Using these estimates and an objective model comparison procedure, the selection proceeds in a forward manner and is controlled by the pre-selected maximum number of covariate terms allowed in the final model. Simultaneous hypothesis testing procedures are used to attempt to asymptotically control the number of unimportant covariates in the final model choice. The methods are evaluated using a simulation study and are applied to a dataset of birth certificate records and infant mortality in the United States.
|
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 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.