Essays on the U.S. Housing Market Dynamics and Boom-Bust Cycles.

摘要

This dissertation includes three essays on theoretical and empirical investigations into the U.S. housing market. Chapter 2 is to extend the dynamic relationship that exists between the US industrial production index, monthly national averages of mortgage loan amount and purchasing price indices for single family houses between the years of 1973 to 2012. The study is achieved by conducting both linear and nonlinear causality tests. For this purpose Vector Autoregression (VAR) approach is used, which is based on error correction model (ECM). We also apply a nonparametric test for Granger non-causality by Diks and Panchenko (2005, 2006) as well as the conventional linear Granger test on the return time series. To ensure that any causality is strictly nonlinear in nature, we also examine the nonlinear causal relationships of pairwise VAR filtered residuals. We find remaining significant bi and uni-directional causal nonlinear relationships in the return series. Chapter 3 is to investigate how the correlations between the U.S. housing market, the credit market and real GDP in the recession of 2007-2009 show different characteristics from the 2001 recession. To do this, we apply Stock and Watson's (1989, 1991, 1993) dynamic factor model with the Kalman filter to construct a coincident factor of the housing market and house credit market separately. We find that the house market and the credit market have strong relationships with real GDP. However, these relationships show some different characteristics for the house credit market in the latest 2007-09 recession from 2001 recession. Chapter 4 focuses on housing cycles, interest rate cycles and business cycles. A nonlinear two-state Markov switching model is used to obtain different regimes in housing cycles, in interest rate cycles and in business cycles. Smoothing probabilities for each cycle is established and, lastly, Filardo's (1994) time varying transition probability model is used to test whether each housing composite measure helps contribute to interest rate and business cycle turning points. The results indicate that the house and credit market common factors contribute to whether the economy remains in an expansionary state or moves into recession and in and high or low interest rate regime. The state of the housing and credit markets contain significant explanatory power for GDP and interest rate fluctuations.