Title of Paper

This paper presents an exploratory simulation experiment to investigate the distribution of coefficients that are used to construct eigenvector spatial filters (ESFs) (Griffith 2000; Griffith 2003). The fundamental idea of ESF exploits the decomposition of a spatial variable into the following three components: trend, spatially structured random component (i.e., spatial stochastic signal), and random noise. The spatially structured component is modeled for with a linear combination of synthetic proxy variables, which are extracted as eigenvectors from a spatial weights matrix that ties geographic objects together in space; it is added as a control variable to a spatial model specification. This control variable identifies and isolates the stochastic spatial dependencies among the georeferenced observations, thus allowing model building to proceed as if the observations are independent. Identifying the distribution of coefficients used to construct this linear combination may support simulation experiments employing the generation of spatially autocorrelated random numbers using ESF.