Testing the Null Hypothesis of Stationarity in Fractionally Integrated Models
28 Pages Posted: 5 Jan 2011
Date Written: December 1999
Abstract
In this paper, we show how to estimate consistently the degree of fractional integration at a given frequency θ, for both stationary and non stationary long-memory process. The statistics used are the periodogram for values θn which converge to θ with an appropriate rate. We also introduce tests of the hypothesis of stationarity for such processes.
Keywords: Fractional integration, Long memory parameter, Spectral density, Moving average unit root, Non parametric tests
JEL Classification: C22
Suggested Citation: Suggested Citation
Do you have a job opening that you would like to promote on SSRN?
Recommended Papers
-
Optimal Supervisory Policies and Depositor-Preference Laws
By Henri Pagès and João A. C. Santos
-
Declining Valuations and Equilibrium Bidding Central Bank Refinancing Operations
By Christian Ewerhart, Nuno Cassola, ...
-
Fiscal Policy in the Transition to Monetary Union: A Structural VAR Model
-
Testing for Zeros in the Spectrum of an Univariate Stationary Process: Part I
-
The Tail Behavior of Stock Returns: Emerging Versus Mature Markets
By Eric Jondeau and Michael Rockinger
-
Inflation and the Markup in the Euro Area
By Catherine Bruneau, Olivier De Bandt, ...
-
Forecasting Inflation in the Euro Area
By Catherine Bruneau, Olivier De Bandt, ...
-
Long-Run Causality, with an Application to International Links between Long-Term Interest Rates
By Catherine Bruneau and Eric Jondeau
-
Assessing GMM Estimates of the Federal Reserve Reaction Function
By Clementine Florens, Eric Jondeau, ...