Small-area estimation using adjustment by covariates

Linear regression models with random effects are applied to estimating the population means of indirectly measured variables in small areas. The proposed method, a hybrid with design- and model-based elements, takes account of the area-level variation and of the uncertainty about the fitted regressi...

Full description

Bibliographic Details
Author: Longford, Nicholas T.
Format: article
Publication Date:1996
Country:España
Institution:Universitat Politècnica de Catalunya (UPC)
Repository:UPCommons. Portal del coneixement obert de la UPC
Language:English
OAI Identifier:oai:upcommons.upc.edu:2099/4066
Online Access:https://hdl.handle.net/2099/4066
Access Level:Open access
Keyword:Statistics
Effective sample size
Linear regression
Random effect
Sampling variation
Mostreig (Estadística)
Classificació AMS::62 Statistics::62D05 Sampling theory, sample surveys
Description
Summary:Linear regression models with random effects are applied to estimating the population means of indirectly measured variables in small areas. The proposed method, a hybrid with design- and model-based elements, takes account of the area-level variation and of the uncertainty about the fitted regression model and the area-level population means of the covariates. The method is illustrated on data from the U.S. Department of Labor Literacy Surveys and is informally validated on two states, Mississippi and Oregon, for which statewide surveys have been conducted.