|
Activity Number:
|
64
- Nonparametric Modeling of Survey Data
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, July 30, 2017 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Survey Research Methods Section
|
|
Abstract #324067
|
|
|
Title:
|
B-Spline Imputation Procedures for the Treatment of Item Nonresponse in Surveys
|
|
Author(s):
|
David Haziza* and Camelia Goga
|
|
Companies:
|
University of Montreal and Universite de Bourgogne
|
|
Keywords:
|
Item nonresponse ;
Nonparametric imputation ;
Survey data ;
Variance estimation
|
|
Abstract:
|
Single imputation is widely used in surveys for treating item nonresponse. Commonly encountered imputation procedures used in practice include parametric procedures such as deterministic and random regression imputation, ratio imputation and random hot-deck imputation. In this presentation we study the properties of imputed estimators obtained under B-spline imputation procedures and derive consistent variance estimators. The results of a simulation study comparing the performance of several imputation procedures in terms of bias and efficiency will be presented.
|
Authors who are presenting talks have a * after their name.
