How GIS can put urban economic analysis on the map

Geographic information systems (GIS) provide powerful tools for storing and manipulating large amounts of information on spatial relationships. However, applications of GIS in urban economics and real estate are infrequent. As a symptom of this, there is often little relation between the extensive literature on residential mortgages and the housing markets literature. The purpose of this paper is to discuss the potential contributions GIS could make to urban economic and real estate economic research and open debate on GIS as applied to these fields, Beginning with theory. we determine how GIS can be useful to researchers and what is needed from future GIS. A GIS is a database management system with a spatial reference. It is capable of providing researchers with excellent control over spatial relationships. For example, one can "geocode" the housing transactions used for hedonic pricing studies. The process of geocoding assigns a latitude and longitude coordinate for each transaction. This enables much more elaborate spatial analysis than with traditional techniques such as using a ruler on a paper map. For instance, GIS allows the hedonic pricing literature to be reworked using spatial autoregressive analysis (SAR). SAR models need to repeatedly calculate large numbers of distances from each observation to every other observation within a given radius. Tn this context, GIS software is primarily useful for assigning a latitude and longitude location to each observation based on the best of street, ZIP, or ZIP+4. These geocoded observations may then be exported to other software for batch processing of SAR statistics. Future development of GIS software should eliminate the need for multiple software packages. Likewise, road distances with flexible weights (e.g., depending on type of road or on construction delays) awaits further software development. (C) 1997 Academic Press.