Author : Yohei Yamamoto
Release : 2009
Genre :
Kind : eBook
Book Rating : /5 ( reviews)

GET EBOOK

Book Synopsis Econometric Essays on Structural Change and Factor Models with Macroeconomic Applications by : Yohei Yamamoto

Download or read book Econometric Essays on Structural Change and Factor Models with Macroeconomic Applications written by Yohei Yamamoto. This book was released on 2009. Available in PDF, EPUB and Kindle. Book excerpt: Abstract: Structural break and factor models have recently been active research areas in time series econometrics. Over the last fifteen years, growing attention has been paid to estimating and testing for multiple structural changes with unknown change points in both theoretical and applied research. Also, estimation and inferential methods for factor models with many observations have received increasing attention. In this dissertation, I make further theoretical contributions in these areas and present empirical macroeconomic applications. In chapter one, I re-examine the relevance of asymptotic optimality criteria in the context of testing for structural break. I critically evaluate Elliott and Müller (2006), who derived a so-called optimal test against the alternative of general parameter variation. However, their framework is based on the assumption that the variation of the parameter process goes to zero at a fast rate. As documented by Kim and Perron (2007), the relative asymptotic efficiency among structural break tests can be different when using the approximate Bahadur slope as the efficiency criterion. Using this measure, I show that the simple Sup-Wald test dominates the Elliott and Müller (2006) test, both in terms of the asymptotic Bahadur efficiency and the finite sample performance. In chapter two, I consider the problem of estimating and testing multiple structural breaks in linear models with endogenous regressors. Based on the results of Perron and Qu (2006), I provide a concise proof of the consistency of the break date estimates obtained via the instrumental variable (IV) method. More importantly, I show that using a method based on ordinary least squares leads to more efficient estimates and more powerful tests compared to using IV procedures when endogeneity is present. In chapter three, I investigate inference problems in factor models in the context of factor-augmented vector autoregressions (FAVAR). I consider identification problems in models with latent processes and propose a bootstrap procedure with factor re-estimation. It is shown to provide improvements over the commonly used bootstrap method suggested by Bernanke, Boivin and Eliasz (2005).