An Income Disaggregated Model of Urban Spatial Structure

Abstract

<p>The task of this paper is to evaluate, in some detail, the effects of income disaggregation upon the predictive and descriptive abilities of a comprehensive, information-minimising, spatial interaction model. The model is comprehensive in that it predicts the probability that an individual household chooses a particular residential location, work place and shopping trip pattern subject to expectations based upon survey data about the average cost of shopping and work trips and the average number of shopping trips per week per household. This probability is information minimising compared to a prior probability distribution which is chosen to reflect residential land availability.</p> <p>The purpose of the analysis is to determine whether or not the income group disaggregated model provides a more accurate representation of observed spatial structure than does the aggregate model (treating the sample data as a homogeneous group).</p> <p>It is concluded that although some descriptive advantages accrue to the disaggregated model, there is generally little predictive advantage to be gained by income group disaggregation unless the specific aim of the study is to discern the differences the exist in locational and trip making behaviour between income categories.</p>