Small area estimation under spatial nonstationarity

Chandra, Hukum, Salvati, Nicola, Chambers, Ray and Tzavidis, Nikos (2012) Small area estimation under spatial nonstationarity. [in special issue: Small Area Estimation] Computational Statistics & Data Analysis, 56, (10), 2875-2888. (doi:10.1016/j.csda.2012.02.006).


[img] PDF - Version of Record
Restricted to internal admin

Download (456Kb) | Request a copy


A geographical weighted empirical best linear unbiased predictor (GWEBLUP) for a small area average is proposed, and an estimator of its conditional mean squared error is developed. The popular empirical best linear unbiased predictor under the linear mixed model is obtained as a special case of the GWEBLUP. Empirical results using both model-based and design-based simulations, with the latter based on two real data sets, show that the GWEBLUP predictor can lead to efficiency gains when spatial nonstationarity is present in the data. A practical gain from using the GWEBLUP is in small area estimation for out of sample areas. In this case the efficient use of geographical information can potentially improve upon conventional synthetic estimation.

Item Type: Article
Digital Object Identifier (DOI): doi:10.1016/j.csda.2012.02.006
Additional Information: 3rd Special Issue on Optimization Heuristics in Estimation and Modelling Problems
ISSNs: 0167-9473 (print)
Keywords: borrowing strength over space, geographical weighted regression, out of sample small area estimation, spatial analysis
Subjects: G Geography. Anthropology. Recreation > G Geography (General)
H Social Sciences > HA Statistics
Divisions : University Structure - Pre August 2011 > School of Social Sciences > Social Statistics
ePrint ID: 181967
Accepted Date and Publication Date:
October 2012Published
15 February 2012Made publicly available
2012In press
December 2011Submitted
Date Deposited: 27 Apr 2011 14:42
Last Modified: 31 Mar 2016 13:36

Actions (login required)

View Item View Item

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