Activity Number:
|
256
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 13, 2002 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Survey Research Methods*
|
Abstract - #300661 |
Title:
|
Inference on Population Means under Unweighted Imputation for Missing Survey Data
|
Author(s):
|
David Haziza*+ and J. N. Rao
|
Affiliation(s):
|
Statistics Canada and Carleton University
|
Address:
|
, Ottawa, Ontario, K1A 0T6, Canada
|
Keywords:
|
Unweighted imputation ; random hot-deck imputation ; uniform response ; ignorable response
|
Abstract:
|
Probability proportional-to-size sampling is often used in surveys, especially for the selection of primary sampling units (clusters). Also, in the presence of item nonreponse, unweighted imputation methods are often used in practice because the edit and imputation stages generally precede the estimation stage so that the sampling weights are not available at the imputation stage. Unweighted imputation methods generally lead to biased estimators under uniform response. Following Skinner and Rao (1999), we propose a bias-adjusted estimator of a population mean under unweighted ratio and random imputation and derive consistent variance estimators. A small simulation study is conducted to study the performance of the proposed estimator in terms of bias and mean squared error.
|