Modelling Locational Factors Using Geographic Information System

MODELLING LOCATIONAL FACTORS USING GEOGRAPHIC INFORMATION SYSTEM GENERATED VALUE RESPONSE SURFACE TECHNIQUES TO EXPLAIN AND PREDICT RESIDENTIAL PROPERTY VALUES

Abdul Hamid b. Hj. Mar Iman al-Murshid

Centre for Real Estate Studies

Faculty of Engineering and Geoinformation Science

Universiti Teknologi Malaysia, Skudai

ABSTRACT

Locational value residual surface (LVRS) techniques can be suggested as an altemative to resolving the difficulty in the traditional modelling of locational influence on property values in a particular area. The VRS can be generated using spatial interpolation techniques such as Inverse Distance Weighting (IDW) and kriging within a Geographical Information System (GIS). It is then incorporated into a regression model as a locational value adjustment factor. The objective of this paper is to compare the relative performance of models that apply LVRS and the traditional multiple regression models in the prediction of residential property values. A controlled sample of single and double-storey residential properties was used to construct regression models. Regressions were run in two stages. The first-stage regression excluded locational component. The differences between the actual and predicted values were then used to construct value residual surface using IDW and kriging techniques. This surface was then used to create locational component that was included in the second-stage regressions. It was found that models applying LVRS were marginally better than the traditional models in predicting property values Besides, the LVRS has allowed for more a visualised locational influence to be evaluated and captured at any geographic points before being statistically modelled in a more effective way.

Keywords: Locational value residual surface, geographical information system, multiple regression analysis, residential property values)