Abstract
A model is presented with which one may identify appropriate spatial lags in a recursive system. The example system represents economic well-being of municipalities in an urban region; the Boston region is used for the case study. The additional explained variation due to the effect of nearby municipalities on the municipality in question may be accounted for with this model. Methods for handling the problems of data additivity and overlap of spatial units are given. The criterion for spatial lag identification is the absolute sum of the path coefficients. Results indicate that poverty levels in a given municipality are affected maximally by an economic variable representing employment and unemployment in neighboring muncipalities up to a distance of eight miles from the center of the community.

This publication has 2 references indexed in Scilit: