This dissertation is composed of three essays on theoretical and empirical investigations into the U.S. housing market. Chapter 1 is to extend a two-sector dynamic general equilibrium model from Ravn, Grohe, and Uribe (2006) deep-habit framework to study asset pricing, particularly housing price. It allows interaction between nondurable and housing markets on both the supply-side and the demand-side. Due to the interaction between the two sectors, the parameter change in the single sector can affect markups of both goods at the steady-state. The findings indicate that the interaction between these two sectors reflects important cross-sector roles of these influential factors in explaining markups.

Chapter 2 is to investigate the seeds of the recent housing and economic crisis by investigating the interrelationship between the recent bust in the housing market, economic recession, and related monetary policies undertaken by the Federal Reserve. A bi-factor dynamic factor model is developed to represent the housing market and the business cycles, and they are allowed to follow different two-state Markov-switching processes. They are linked through monetary policy as reflected in interest rates movements. We find a strong correlation between the business cycle and interest rates, and between the housing cycle and interest rates, and between the two cycles. However, the close relationship changes since the 2001 recession as interest rates end its low-level due to the uncertainty and the slow recovery after this recession. The missed linkage between the business cycle and the housing cycle worked as a seed of the housing market bust and the economic recession in 2007-2009.

Chapter 3 aims to investigate the effect of a potential important driver of the recent housing price boom and bust - people's expectations on the U.S. housing asset returns. Particularly, it extends the volatility feedback model proposed in Kim, Morley and Nelson (KMN 2004) to study the relationship between housing volatility and returns during 1963-2007. The analysis considers two alternative breakpoints--1984Q1 and 1999Q1-- to distinguish permanent structural breakpoints from Markov-switching. The results indicate that the relationship between the U.S. housing volatility and the expected returns is significantly positive. Thus, the important role of people's expectation is strongly supported. The current U.S. housing bubble can be explained by the relationship between housing volatility and realized returns, and that between housing volatility and expected returns. Corresponding to Chapter 2, this chapter also indicates a strong association between housing cycles and business cycles, and a remarkable uncertainty in the U.S. housing market during the post-1999.