This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Permalink
https://hdl.handle.net/2142/85672
Description
Title
Essays on Time Series Modeling
Author(s)
Campelo, Ana Katarina
Issue Date
2000
Doctoral Committee Chair(s)
Anil Bera
Department of Study
Economics
Discipline
Economics
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
Economics, Theory
Language
eng
Abstract
The first chapter of this dissertation introduces a few important concepts which are used throughout the rest of the dissertation. Chapter 2 revisits the inertial inflation hypothesis for Brazil. In this chapter we use time series techniques to model the long-run dynamics of the Brazilian inflationary process. Our results reveal that, although there is some inertia in the Brazilian inflation, the degree of inertia is rather small. Another important policy issue conjectured by Friedman in his Nobel Lecture is the relationship between inflation and uncertainty about the future level of inflation. The third chapter uses quantile regression techniques to test Friedman's hypothesis using the U.S. implicit price deflator for GNP as a measure for inflation. We find some evidence in favor to Friedman's claim that there exists a positive and significant relationship between the inflation level and inflation uncertainty (measured here as the conditional scale of inflation). However, it is shown that this relationship only holds for certain ranges (higher quantiles) of the conditional quantiles of our measure of scale. Finally, the fourth chapter evaluates the predictive accuracy of interval forecasting methods for ARCH-type models proposed in the literature under different misspecification scenarios: misspecification of the conditional distribution, misspecification of the conditional variance, and misspecification of both the conditional variance and the distribution. Our first main result is that, overall, the bootstrap method outperforms the alternatives, yielding prediction intervals whose empirical levels and nominal levels match well, even under a misspecified conditional variance. Our second important result in this chapter is that the GARCH(1,1) specification is reliable for building prediction intervals which approximate well the nominal confidence levels when the true data generating process comes from a different conditional heteroskedastic model.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
