it would be best if someone lets me know with reference to a publication. because there are many opinions. Also , I know if series are co-integrated we should use VEC model but some paper argues to use var mode to test grangare causality. Could some please also comment on this ?
thank you very much for your time and effort :)
Dear
hope you find this links useful. it contains full description and details of applying Granger test using VAR model. it also include details about structural analysis.
http://faculty.washington.edu/ezivot/econ584/notes/varModels.pdf
a more advance working paper which address cointegrated series and Granger causality can be found at:
http://www.brandeis.edu/departments/economics/RePEc/brd/doc/Brandeis_WP36.pdf
regards;
Amer.
I am not an expert. But first you have to test the stationarity of the series for the VAR to be valid. For macroeconomic series, the level is usually not stationary, but the first difference, or the growth rate, is.
http://faculty.washington.edu/ezivot/econ584/stck_watson_var.pdf
As far as I am aware, always first difference. Granger is short term test!
- Badges
- Science topic
- Similar topics
- Biological Science
- Bioecology
- Systems Ecology
- Landscape Ecology
- Connectivity