Development and Application of Bayesian Nonparametric Methods in Time Series Analysis

Abstract

Motivated by applications of time series analysis in economics, I adopt Bayesian nonparametric methods to build novel econometric methodologies. In three self-contained chapters, this thesis demonstrates how these proposed modelling frameworks can be applied to extending knowledge in the corresponding empirical studies. Chapter 2 exploits data on a set of traded goods to undertake the first comprehensive empirical analysis of market integration between the Ottoman Empire and Europe from 1469 to 1914. Computing dynamic factor models via Bayesian inference, we overcome such data constraints as missing observations and a small sample size. The results of this analysis suggest that there were p..