Tufts University This research proposal focuses upon spatial aspects of economies with interacting agents. A hybrid approach is sought to construct models which combine fixed and a random topology of interactions. It is argued that in depth studies of circular patterns of interaction are crucial for understanding diverse issues such as micro aspects of social structure and the emergence of monetary economies. The model development emphasizes two empirical applications. First, a study of urban neighborhood dynamics in U.S. cities is proposed. Second, a study of the spatial aspects of the evolutionary behavior of the U.S. system of cities will be examined. The neighborhood example will focus on the dynamics of residential segregation and equality as a way of understanding the dynamics of webs of interaction which constitute the structure of the urban economy. PSID and AHS data are used within a spatial econometric framework to derive generalizable results and insights. This analytical approach also allows the researchers to link their work to interacting agent theory.
