Modeling Firm Demography in Urban Areas with an Application to Hamilton, Ontario: Towards an Agent-Based Microsimulation Model

Abstract

<p>The concern about sustainable planning of cities over the past few decades has generated a considerable amount of empirical literature on the emerging urban form and its causes. Part of this literature has focused on modeling the behavior driving the location decision of households and firms, as well as the urban development practices to simulate the evolution trends of urban form and growth. Earlier efforts were based on building and using aggregated zone-based urban land use models. However, due to their aggregate nature, most of these models have been criticized for lacking the behavioral realism in imitating the complexities of the city development process. Research during the past decade and a half calls for the adoption of a micro-analytical approach, whereby change in urban form is driven by the collective behavior of individual agents (households, business establishments, developers, etc.) and their location decisions.</p> <p>This dissertation is concerned with the behavior of business establishments in urban areas. Specifically, we use principles from firm demography to develop a framework for an agent-based microsimulation model for the City of Hamilton, Ontario, Canada. The framework will be used to simulate the evolution of small and medium size business establishments as an outcome of processes that related to the failure, mobility and location choice of establishments in the city. The statistical models underlying these processes are discussed and estimated using data extracted from Statistics Canada Business Register (BR) for the period 1996 -2002. The parameters of these models attempt to identify and quantify the causes associated with firmographic processes in Hamilton and will form the basis towards implementing an agent-based model.</p>