Résumé 185 :

Accuracy of areal interpolation methods: count data
Do, Van Huyen ; THOMAS AGNAN, Christine ; VANHEMS, Anne
GREMAQ Université Toulouse 1 - Capitole

The analysis of socio-economic data often implies the combination of data bases originating from different administrative sources so that data have been collected on several different partitions of the zone of interest into administrative units. It is therefore necessary to allocate the data from the source spatial units to the target spatial units. A particular case of that problem is when one uses a common regular grid and re-allocate everything to this single target partition. This option is currently under study in France at INSEE and in Europe with the EU directive 'INSPIRE', or INfrastructure for SPatial InfoRmation. There are three main types of such techniques: proportional weighting schemes, smoothing techniques and regression based interpolation. We propose a model based on Poisson point patterns to consider the accuracy of these techniques for regular grid targets in the case of count data. The error depends on the nature of variable of interest and its correlation with the auxiliary variable. We focus on proportional weighting schemes and Poisson regression based methods. There is no technique which always dominates.