Using the Bootstrap technique and the sample autocorrelation function for the identification of unit root processes

>By Panagiotis Mantalos .

Abstract: By using a Bootstrap procedure and the function of sample autocorrelations of a time series, we are able to distinguish between a non-stationary integrated series of order one, I(1), from a stationary AR(1) process. We investigated our Bootstrap procedure by using a Monte Carlo method and by an empirical example. The results show that by using the Bootstrap average of the sample autocorrelations function, we can easily identify a non-stationary I(1).

Key Words: Bootstrap, Co-integration, Sample Autocorrelations, Unit Root

Author:
Panagiotis Mantalos, panagiotis.mantalos@bth.se

Editor: Rob J. Hyndman , Rob.Hyndman@monash.edu.au

READING THE ARTICLE: You can read the article in portable document (.pdf) format (377379 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 2122 times since July 24, 2006.


Return to the InterStat Home Page.