TESTING AREAL INTERPOLATION METHODS WITH US CENSUS 2010 DATA

Huyen DO van, Christine Thomas-Agnan and Anne Vanhems ()
Additional contact information
Huyen DO van: Toulouse School of Economics (GREMAQ)

Region et Developpement, 2014, vol. 40, 83-96

Abstract: The areal interpolation problem is that of projecting a characteristic of interest on a partition of space called target partition from the knowledge of the same characteristic on a different partition, so called source partition, using some auxiliary information. The objective of this paper is to use a demographic database available in the R package ‘US census 2010’ (Almquist, 2010) in order to test several areal interpolation methods based on regression in the case of count related data. The fact that data is available at many different spatial scales in this database make this comparison study unique. Another innovative point of view is that we compare the extensive approach versus the intensive approach for a variable which is a ratio of counts. We also include the comparison with the scaled regression for the extensive case introduced in Do et al. (2015) and with a scaled regression for the intensive case proposed here. Finally we give some empirical guidelines for the choice of auxiliary information.

Keywords: AREAL INTERPOLATION; SPATIAL DISAGGREGATION; PYCNOPHYLACTIC PROPERTY (search for similar items in EconPapers)
JEL-codes: C15 C83 R15 (search for similar items in EconPapers)
Date: 2014
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://regionetdeveloppement.univ-tln.fr/wp-content/uploads/6_DoVan.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:tou:journl:v:40:y:2014:p:83-96

Access Statistics for this article

More articles in Region et Developpement from Region et Developpement, LEAD, Universite du Sud - Toulon Var Contact information at EDIRC.
Bibliographic data for series maintained by Christophe Van Huffel ().