Date of Award

May 2018

Degree Type


Degree Name

Doctor of Philosophy (PhD)




Stuart S. Rosenthal


Demand shocks, Housing dynamics, Housing supply, Spatial dependence

Subject Categories

Social and Behavioral Sciences


This dissertation consists of two essays on housing market dynamics and cointegration analysis with latent factors. The theme of this dissertation is housing market dynamics, with the first essay an application of advanced panel time series models to the studies of housing market dynamics, and the second essay a theoretic derivation of an econometric tool on cointegration analysis with latent factors that can be applied to the housing market analysis.

This first essay develops a parsimonious dynamic model to study the impact of a common demand shock in the housing market on housing prices and construction activities across a set of locations with heterogeneous supply side conditions. Embedded within the model is a lead-lag structure that allows one to identify from where shocks propagate while allowing for and yielding estimates of cross-sectional differences in housing supply elasticities. The findings indicate that local supply conditions may matter more than distance when modeling spatiotemporal dynamics in the housing market.

The second essay considers estimating and testing cointegration between an integrated series of interest and a vector of possibly cointegrated nonstationary latent factors. The fully modified least squares (FM-OLS) estimation is adopted to the estimation of the cointegration relation of interest. The asymptotic properties of the FM-OLS estimators are derived, and the residual-based cointegration tests are shown to work as usual even with latent factors. Based on the estimated cointegration relation, it is demonstrated that an error correction term added to the traditional diffusion index forecast model improves forecasting accuracy.


Open Access