-gcause-, a module to perform Granger causality tests, is now available on
SSC. An abstract follows.
Users are advised to use -gcause- rather than the -granger- package from
STB51[sts14] as the latter contains numerous errors in both its syntax and
formulas. Users of the -granger- package might not notice the
inconsistencies in the test statistics since the program will not
necessarily fail to execute. An attempt was made to contact the author of
granger.ado, J.Sky David, to suggest amendments to the original code. These
attempts were not successful.
Thanks to Kit Baum for placing this module on the archive and to Carol Miu
for suggesting an enhancement to the code. Thank you also to Nick Cox for
inquiring to Prof. Joe Newton about the status of J.S. David's affiliation
with the Texas A&M political science dept.
TITLE
'GCAUSE': module to perform Granger causality tests
DESCRIPTION/AUTHOR(S)
gcause performs a Granger causality test on time-series data to
investigate whether lagged values of one variable help in
forecasting another variable. This routine should be preferred to
an earlier, problematic Granger-test routine which appeared in
STB-51.
Distribution-Date: 20021009
Author: Patrick Joly, Industry Canada
Support: email [email protected]
INSTALLATION FILES (type net install gcause)
gcause.ado
gcause.hlp
Patrick Joly
Industrial Analysis Centre, 730E
941-3327
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
