|
Activity Number:
|
668
- Estimation with Statistical Models
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 3, 2017 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Survey Research Methods Section
|
|
Abstract #324286
|
View Presentation
|
|
Title:
|
Semiparametric Adaptive Estimation with Nonignorable Nonresponse Data
|
|
Author(s):
|
Kosuke Morikawa* and Jae-kwang Kim
|
|
Companies:
|
Osaka University and Iowa State University
|
|
Keywords:
|
Estimating functions ;
Identification ;
Incomplete data ;
Not missing at random (NMAR) ;
Semiparametric Efficient Estimation
|
|
Abstract:
|
When the response mechanism is believed to be nonignorable or not missing at random (NMAR), a valid analysis requires stronger assumptions about the data than do standard statistical methods. Semiparametric estimators have been developed under the correct model specification assumption for the response mechanism. In this talk, we consider a scheme for obtaining the optimal estimation for the parameters such as the mean and propose two semiparametric adaptive estimators that do not require any model assumptions except for the response mechanism. Asymptotic properties of proposed estimators are discussed.
|
Authors who are presenting talks have a * after their name.
