Imputation using higher order moments of an auxiliary variable

In this article, we propose a new method of imputation that makes use of higher order moments of an auxiliary variable while imputing missing values. The mean, ratio, and regression methods of imputation are shown to be special cases and less efficient than the newly developed method of imputation,...

Full description

Saved in:
Bibliographic Details
Published in:Communications in statistics. Simulation and computation 2017-09, Vol.46 (8), p.6588-6617
Main Authors: Mohamed, Choukri, Sedory, Stephen A., Singh, Sarjinder
Format: Article
Language:English
Subjects:
Auxiliary information
Bias and variance
Imputation
Mathematical analysis
Missing data
Regression analysis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this article, we propose a new method of imputation that makes use of higher order moments of an auxiliary variable while imputing missing values. The mean, ratio, and regression methods of imputation are shown to be special cases and less efficient than the newly developed method of imputation, which makes use of higher order moments. Analytical comparisons show that the first-order mean squared error approximation for the proposed new method of imputation is always smaller than that for the regression method of imputation. At the end, the proposed higher order moments-based imputation method has been applied to a real dataset.
ISSN:0361-0918
1532-4141
DOI:10.1080/03610918.2016.1208235