Advances in integrated urban modeling: Microsimulation models of the housing market, real estate development, and workplace choice

This dissertation is comprised of three papers advancing the current practice of integrated urban modeling.The first paper addresses previous limitations of the Random Biding Model (RBM) and adapts it for microsimulating housing markets. The RBM developed by Ellickson and Lerman and Kern enables an empirical, discrete choice interpretation of the classic bid rent theory presents an alternative to estimate willingness-to-pay for housing characteristics and models housing choices along with transaction prices in short-run microsimulations of housing markets. We address the limitations of RBM by proposing an individual household level RBM with endogenous screening rules. A Bayesian approach is adapted that enables us to estimate the RBM with alternative distribution of the error term, as well as recover the screening rules along with the choice model simultaneously from revealed preference data.The second paper models the real estate development process, adding a novel alternative to the two traditional approaches, site transition and project siting, respectively. We develop a model that is based on economic theory of developer behavior which can be reconciled with these previous approaches, and apply it to the Central Puget Sound region of Washington State. We model real estate development through a probabilistic profit maximization process at the parcel level, which allows our model to consider competition for both development sites and real estate product types at more realistic units of analysis than previous models.The final paper enhances the integration of land use and transportation modeling by proposing a replacement of the home-based work destination choice model within the 4-step travel model system with a pair of choice models at the individual worker level. These two choice models predict whether a worker will choose to work at home on a long-term basis, and if not, then choose an out-of-home job. These models link an individual worker to a specific job at a workplace, and therefore directly predict commuting patterns. We present the model specification, estimation results, and results of validating the models against observed commuting data from the Census Transportation Planning Package.