Spatial Clustering Methods and Small Area Estimation

Local statistical offices often dispose of very rich databases of spatially referenced socio– economic data. The high degree of spatial detail of such information is often not too useful for practical purposes in that firms or local authorities are interested in information aggregated at higher levels. The standard practice usually consists in aggregating the data at some prespecified geographical level (say, city districts). A more statistically sound approach consists in let the data speak for themselves in order to identify spatial clusters from the data. This work presents methods for the definition of boundaries of spatially referenced socio economic phenomena. The area definition and the assignment of the data to appropriate areas can pose problems in the estimation process. In particular, in small area estimation the importance of this matter is represented by the fact that some parameters of the model can be related to the between-area relationships.