Essays on time series models with dynamic coefficients in macroeconomics and finance.

摘要

With a growing body of recent empirical evidence on widespread instability in macroeconomic time series relations (Stock and Watson, 1996), Markov-switching models and time-varying parameter models have drawn more attention than ever. In my dissertation, various issues related to these models are considered with applications to macroeconomics and finance. In the first essay, I apply a regime-switching model with endogenous regressors to investigate the nature of structural changes in the hybrid New Keynesian Phillips curve (NKPC). The empirical results show multiple structural breaks in the NKPC in 1974 and 1982. Accounting for these structural changes, the backward-looking component is no longer significant throughout the whole sample period. In the second essay, I provide a framework for dealing with the endogeneity problem in the time-varying parameter models. Both joint estimation and two-step estimation procedures are derived. In my third essay, I develop an empirical model of stock returns and economic activity that allows one to examine the business conditions-risk-return relationship. In doing so, I consider the possibility that stock market volatility may increase during recessions and other short periods of liquidity crisis such as the 1987 stock market crash and the 1998 Russian default. Empirical results show that business conditions-related market volatility has predictive power for the expected return movements, while business conditions-unrelated volatility does not.