|
Activity Number:
|
193
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 4, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #302707 |
|
Title:
|
Multiple imputation in multiple classification and multiple-membership structures
|
|
Author(s):
|
Recai M. Yucel*+
|
|
Companies:
|
SUNY-Albany
|
|
Address:
|
Department of Epidemiology and Biostatistics, Rensselaer, NY, 12144-3456,
|
|
Keywords:
|
multiple imputation ; Bayesian inference ; missing data ; multiple membership ; mixed-effects ; administrative data
|
|
Abstract:
|
In data systems with complexities due to nested/non-nested clustering and multiple-membership, missing values present an added analytic challenge to the statistical analyses. We develop model-based multiple imputation (MI) inference which has been a popular method in the analyses of missing data. Adaptations of multivariate generalizations of the mixed-effects models are used as imputation model. These models are modified to handle multivariate responses and observational units with possibly overlapping membership of clusters that are not necessarily hierarchical. Markov Chain Monte Carlo techniques are used to simulate and draw imputations from underlying joint posterior predictive distributions. Relevant concepts on both multiple-membership and non-nested clustering are demonstrated longitudinal administrative data with panel missingness as well as arbitrary item nonresponse.
|
