Modelling complex survey data with population level information: an empirical likelihood approach
Modelling complex survey data with population level information: an empirical likelihood approach
Survey data are often collected with unequal probabilities from a stratified population. In many modelling situations, the parameter of interest is a subset of a set of parameters, with the others treated as nuisance parameters. We show that in this situation the empirical likelihood ratio statistic follows a chi-squared distribution asymptotically, under stratified single and multi-stage unequal probability sampling, with negligible sampling fractions. Simulation studies show that the empirical likelihood confidence interval may achieve better coverages and has more balanced tail error rates than standard approaches, which involve variance estimation, linearization or resampling.
design-based inference, empirical likelihood, estimating equation, inclusion probability, regression parameter, unequal probability sampling
447-459
Oguz Alper, Melike
d555d994-86fa-4835-80da-c5a3b0a91bcc
Berger, Yves G.
8fd6af5c-31e6-4130-8b53-90910bf2f43b
Oguz Alper, Melike
d555d994-86fa-4835-80da-c5a3b0a91bcc
Berger, Yves G.
8fd6af5c-31e6-4130-8b53-90910bf2f43b
Oguz Alper, Melike and Berger, Yves G.
(2016)
Modelling complex survey data with population level information: an empirical likelihood approach.
Biometrika, 2 (103), .
(doi:10.1093/biomet/asw014).
Abstract
Survey data are often collected with unequal probabilities from a stratified population. In many modelling situations, the parameter of interest is a subset of a set of parameters, with the others treated as nuisance parameters. We show that in this situation the empirical likelihood ratio statistic follows a chi-squared distribution asymptotically, under stratified single and multi-stage unequal probability sampling, with negligible sampling fractions. Simulation studies show that the empirical likelihood confidence interval may achieve better coverages and has more balanced tail error rates than standard approaches, which involve variance estimation, linearization or resampling.
Text
OguzAlper_Berger_2016
- Accepted Manuscript
Restricted to Repository staff only until 31 May 2017.
Request a copy
More information
Accepted/In Press date: 17 March 2016
e-pub ahead of print date: 23 May 2016
Keywords:
design-based inference, empirical likelihood, estimating equation, inclusion probability, regression parameter, unequal probability sampling
Organisations:
Statistical Sciences Research Institute
Identifiers
Local EPrints ID: 376699
URI: https://eprints.soton.ac.uk/id/eprint/376699
ISSN: 0006-3444
PURE UUID: 1fd9d044-9964-42eb-81f1-7edc75cf116c
Catalogue record
Date deposited: 05 May 2015 14:36
Last modified: 15 Jul 2019 18:26
Export record
Altmetrics
Contributors
Author:
Melike Oguz Alper
University divisions
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics