Essays on international real business cycle models and Bayesian estimation
My dissertation consists of three essays that study the properties of international business cycle, and the underline driving forces of their comovement. In the first essay, I check the determinants of business cycle comovement by using a Bayesian dynamic factor model. I show that the services sector and the industry sector have similar patterns of comovement for most countries. My results suggest that the input-output theory largely explain the business cycles comovement across countries at the sectoral level. The empirical evidence also supports the idea that the value of services is strongly correlated with the value of industrial production. The second essay examines parameter heterogeneity in the small open economy model. I use Bayesian methods to fully estimate the small open economy model augmented with both permanent and transitory productivity shocks for a large group of countries. I find that structural parameters are moderately heterogeneous across countries, and ignoring such heterogeneity will introduce biases into the estimation of productivity shocks, especially for developing countries. The third essay compares the performance of two types of international real business cycle models, namely two-country dynamic stochastic general equilibrium (DSGE) and small open economy (SOE) models in explaining the shock transmission mechanisms across countries. In the two-country DSGE model, comovement of business cycles among different countries can be captured by correlated permanent and transitory components of productivities across countries. But in the SOE model, comovement of economies can only be modeled through the channel of world interest rate shock. Our evidence shows that the two-country DSGE model outperforms the SOE in capturing the effects of the permanent shocks originated from foreign countries, especially for the developed countries, which are highly integrated with each other.