Skip navigation
Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorSims, Christopher
dc.contributor.authorLi, Dake
dc.contributor.otherEconomics Department
dc.description.abstractI develop and apply econometric methods for estimating impulse responses of macroeconomic variables to structural shocks. Chapter 1 uses a structural vector autoregressive model with a Zero Lower Bound to characterize the censored nominal interest rate and the effect of unconventional monetary policy. Under empirically relevant non-Gaussian shocks, this chapter proposes a semiparametric identification scheme to prove point identification, without relying on the parametric form of the shock distribution. The key problem of model identification that this chapter solves is to deal semiparametrically with the non-linearity arising from censoring of the nominal interest rate at zero. An efficient Bayesian inference routine is designed to facilitate model estimation in practice. The empirical results suggest that the unconventional monetary policy has a small and transitory effect. Chapter 2 is a joint work with Mikkel Plagborg-Møller and Christian K. Wolf, where we conduct a simulation study of Local Projection (LP) and Vector Autoregression (VAR) estimators of structural impulse responses across thousands of data generating processes, designed to mimic the properties of the universe of U.S. macroeconomic data. Our analysis considers various identification schemes and several variants of LP and VAR estimators. A clear bias-variance trade-off emerges: LP estimators have lower bias than VAR estimators but substantially higher variance at intermediate and long horizons. Consequently, unless researchers are overwhelmingly concerned with bias, shrinkage via Bayesian VARs or penalized LPs is attractive. Chapter 3 is a joint work with Christopher A. Sims, where we estimate a structural VAR on a balanced panel of 10 countries to examine different causal channels that lead to comovement between aggregate credit and GDP. Among all the six structural shocks in our model, we find that there is one that pushes aggregate credit up and then later depresses GDP growth, but its effects are small in most countries. All the other five structural shocks still generate the positive or zero comovement. The shocks in our model are identified via variation across countries in their relative size, with their global trend taken into account. We also model the dependence between initial conditions and country constants to avoid the small-T panel bias.
dc.publisherPrinceton, NJ : Princeton University
dc.relation.isformatofThe Mudd Manuscript Library retains one bound copy of each dissertation. Search for these copies in the library's main catalog: <a href=></a>
dc.subjectimpulse response
dc.subjecttime series
dc.subjectvector autoregression
dc.titleEssays on Estimating Impulse Responses in Macroeconometrics
dc.typeAcademic dissertations (Ph.D.)
Appears in Collections:Economics

Files in This Item:
File Description SizeFormat 
Li_princeton_0181D_14262.pdf2.57 MBAdobe PDFView/Download

Items in Dataspace are protected by copyright, with all rights reserved, unless otherwise indicated.