Online Program Home
My Program

Abstract Details

Activity Number: 667
Type: Contributed
Date/Time: Thursday, August 4, 2016 : 8:30 AM to 10:20 AM
Sponsor: Section on Nonparametric Statistics
Abstract #319639 View Presentation
Title: Nonparametric Kernel Density Estimation Using Auxiliary Information from Complex Survey Data
Author(s): Sayed A. Mostafa* and Ibrahim A. Ahmad
Companies: Oklahoma State University and Oklahoma State University
Keywords: Kernel Density Estimation ; Complex Survey Data ; Auxiliary Information ; Superpopulation

This paper presents some new and serious attempts towards using auxiliary information effectively in kernel density estimation for data from complex surveys. We develop new kernel density estimators (KDEs) that use both complete auxiliary information and the sample information in the framework of complex surveys. The new KDEs are obtained by modelling the relationship between the study variable and the auxiliary variable using both parametric and nonparametric regression models. Two model-assisted KDEs for the density function of the study variable are proposed. The statistical properties of these estimators are studied under a combined design-model-based inference framework which accounts for both the underling model and the randomization distribution. A global error criterion is used to determine the optimal smoothing parameter for each estimator. Data-driven bandwidth estimators are obtained using plug-in techniques. Using Monte Carlo methods, we address the finite sample properties of the proposed KDEs. Additionally, we compare the new estimators with standard estimators that ignore the auxiliary information.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association