Mixed Unit Roots and Deterministic Trends in Noncausality Tests

Using Japanese economic data and a Monte Carlo simulation, this study analyzes the consequences of ignoring deterministic trends in mixed unit-root data for Granger noncausality tests. Results from an augmented VAR suggest over-rejection in certain empirically relevant cases at various sample sizes.


Issue Date:
2008
Publication Type:
Conference Paper/ Presentation
PURL Identifier:
http://purl.umn.edu/6745
Total Pages:
20




 Record created 2017-04-01, last modified 2017-11-16

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)