|
Activity Number:
|
512
- Various Flavors of Missing-Data Problems
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2018 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Survey Research Methods Section
|
|
Abstract #329247
|
Presentation
|
|
Title:
|
Hybrid Imputation Models Through Blocks
|
|
Author(s):
|
Stef van Buuren*
|
|
Companies:
|
TNO
|
|
Keywords:
|
MICE; missing data; joint modeling; fully conditional specification
|
|
Abstract:
|
Joint modeling and fully conditional specification are two seemingly disparate approaches to multivariate imputation. A block is a collection of variables. The new MICE algorithm iterates over blocks instead of variables. We obtain the old algorithm by specifying each block as one variable, which is still the default. Specifying blocks in other ways produce hybrids of univariate and multivariate imputation methods, some which have interesting applications. This presentation reviews some of the newer imputation models.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2018 program
|
