Spatial Econometric Functions in R

Developments in the R implementation of the S data analysis language are providing new and effective tools needed for writing functions for spatial analysis. The release of an R package for constructing and manipulating spatial weight, and for testing for global and local dependence during 2001 has been followed by work on functions for spatial econometrics (package spdep; the package may be retrieved from: http://cran.r-project.org). This chapter gives an introduction to some of the issues faced in writing this package in R, to the use of classes and object attributes, and to class-based method dispatch. In particular, attention will be paid to the question of how prediction should be understood in relation to the most commonly employed spatial econometrics simultaneous autoregressive models. Prediction is of importance because fitted models may reasonably be expected to be used to provide predictions of the response variable using new data − both attribute and position − that may not have been available when the model was fitted.