I have estimated a VECM in Stata with rank =2 on five variables after confirming that all variables are I(1) and there are 2 CI rank using the Johansen test. Now I want to check CI relations stability using the recursive procedure proposed by Hansen and Johansen (1992, 1999). I do not know how to go about this in Stata or EViews or R.
I did the search for you. On page 307 of the Hansen-Johansen (1999) paper, the authors mention the statistical packages PcGive (Doornik and Hendry, 1994) and CATS in RATS (Hansen and Juselius, 1995). Both are commercial products. I didn't see about recursive information in the former (at least https://www.doornik.com/doc/PcGive/index.html, I didn't look at the whole Ox family) but the latter is still there, see https://www.estima.com/catsinfo.shtml, but requires buying RATS.
Hi Guy, Thank you for your response. I also saw that https://www.econstor.eu/bitstream/10419/26208/1/557976286.PDF used the CATS in RATS to implement the Hansen and Johansen recursive method. I guess I will have to check with my Uni for access to RATS.
IN Stata, use the vecstable command. This command tests for the stability of the cointegrating vectors by recursively estimating the cointegrating vectors over a rolling window of the data .
EViews, useEViews, use the Johansen cointegration test to check for cointegration and the stability of the cointegrating vectors. It can estimate the number of cointegrating vectors and their coefficients.
Yes, you can check the stability of the cointegration relation using the Hasza and Johansen recursive method in EViews.
- Badges
- Science topic