01474941

Component

This paper presents a disaggregated model of real estate demand with price formation for use in integrated land use and transportation modeling. Previous land use models for similar purpose either pursue an aggregated equilibration approach or a disaggregated dis-equilibrium approach without much realistic representation of price movement. We build a disaggregated real estate demand model based on discrete choice model where agents (households) choose residence location based on Random Utility Maximization principle and the prices move to reflect competition among agents. Our model has the advantage over previous models of being able to capture realistic price formation while representing heterogeneous taste of agents in a dynamic modeling framework. Our model has been implemented as a module for UrbanSim, a micro-simulation modeling system for land use, transportation, and environment, allowing users to easily swap between versions of residence location choice models with and without price formation. Implemented in Python and utilizing optimized scientific computing packages numpy and scipy, the module is very efficient and allows models of a large metropolitan area to run on standard computer hardware in reasonable time. In this paper we show the structure and implementation of the model, demonstrate its performance, and present model estimation and sensitivity testing results using housing and population data of the San Francisco Bay Area.

This paper was sponsored by TRB committee ADB40(2) Integrated Transportation and Land Use Modeling.

01470560

13-1718
13-1718

English

12p

2013

Transportation Research Board 92nd Annual Meeting

Digital/other

Figures; References; Tables

Choice models; Demand; Land use models; Location; Microsimulation; Multinomial logits; Pricing; Real property

Data and Information Technology; Highways; I70: Traffic and Transport

Transportation Research Board Annual Meeting 2013 Paper #13-1718

TRIS, TRB, ATRI

Feb 5 2013 12:25PM