Spatial Distribution of Population: Cities and Suburbs*

Model Comparison and Model Validation Issues in Empirical Work on Urban Density Functions

“…In particular, following a suggestion in Hamson and Kain (1974), density functions are estimated for a purer measure of residential density than the conventional tract density, using detailed information on land use. As a result, similar to the approach in Martin (1973), this net density can be compared to the more traditional gross density (measured with tract area-often because of a lack of more specific data) and the extent to which the two implementations agree in an empirical situation can be assessed. Also, in the empirical application, two types of data sets are considered, to reflect the situation of a random sampling of spatial units, as well as the one where contiguous units are used (e.g., when all data in a metropolitan area are available).…”

Urban Population Spacing Analysis

“…They show that the required assumption that the price elasticity of demand for housing is -1 does not obtain about half the time. Earlier, Martin (1973) found considerable differences in structure between cities and suburbs and concluded that a single gradient is an inappropriate representation for entire areas. Odland (1978) studied four middle-size cities and concluded that polycentric urban forms are likely to prevail over the single center form when city size becomes large relative to the marginal costs of commuting.…”