University of Illinois Urbana-Champaign

Experience in the application of unit roots and fractional difference models and tests

Agiakloglou, Christos N.

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/2142/19013

Description

Title
Experience in the application of unit roots and fractional difference models and tests
Author(s)
Agiakloglou, Christos N.
Issue Date
1992
Doctoral Committee Chair(s)
Newbold, Paul
Department of Study
Economics
Discipline
Economics
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
  • Statistics
  • Economics
  • Economic Theory
Language
eng
Abstract
  • One of the most important aspects in analyzing economic time series is to specify whether the observed series is generated by a stationary or non-stationary process, since most macroeconomic variables could be generated by a unit autoregressive root process. This determination as to whether or not a series should be differenced is known as the unit root test. In other words, if a time series in non-stationary, then in the presence of a unit autoregressive root, it can be converted to a stationary and invertible process by first differencing.
  • This thesis examines some techniques for testing the unit root hypothesis. It evaluates first the Dickey-Fuller-Type Tests for a unit autoregressive root in the presence of moving average components for any ARIMA (p, 1, q) process, proposed by Said and Dickey (1984, 1985). This type of test is conducted by either fitting a long autoregression to the data when the orders p and q are unknown, or by using the one-step Gauss-Newton least squares estimation when the orders p and q are known. Furthermore, to investigate the performance of the above tests, some applications of unit root models presented in the recent literature are also re-examined.
  • Moreover, this thesis employs the analysis of fractionally integrated models, which nests the unit root phenomenon, to distinguish in a more general way the value of the difference parameter. First, it evaluates and provides examples of the Geweke and Porter-Hudak (1983) estimation procedure which is frequently used to obtain an estimate of the fractional difference parameter, d. Next, it presents some new alternative testing techniques for fractionally integrated models. Finally, the behavior of the sample autocorrelations of fractional white noise is investigated.
Type of Resource
text
Permalink
http://hdl.handle.net/2142/19013
Copyright and License Information
Copyright 1992 Agiakloglou, Christos N.

Owning Collections

Graduate Dissertations and Theses at Illinois PRIMARY
Graduate Theses and Dissertations at Illinois
Dissertations and Theses - Economics