I am currently dealing with some monthly data on interest rates which goes back over thirty years.
The total 30 years worth of data is I(1). However when I shorten this data to go back only 10 years its I(2).
When running this data in a VECM for the 10 year interval, do I have to difference or log the variable it for it to be I(1) or is it more appropriate just to treat it like an I(1) variable (since the true nature of the process is I(1)) and put it in the VECM even though the 10 years worth of data is I(2)?
Dear Jacob,
It seems inappropriate when the series have the evidence of I(2) and we deal it as I(1), try transformation if you want to use VECM or VAR.
Regards,
Atif
i think you have to be specific as regards your scope, if 10 years series gives I(1) then the use of VECM is appropriate but if its I(2), i will advise you go for GMM
I hope this helps
How are sure that the variable is I(1) or I(2)? What type of/how many unit root tests did you use?
Normally, it is better to base our decision regarding variable order of integration on several unit root tests.
Did you take the potential structural break (i.e., 2008 crisis or any other important event) in the series while testing for unit root. If not, I advise you to use unit root with structural break. The traditional unit root tests produce inaccurate results especially when the break exist in the data. This highly relevant to your case, assuming that 2008/2009 is part of the period covered by your data, since many countries have adjusted their policies (monetary) after the crisis. Therefore, testing for unit root while accounting for such break in the series is critically important.
- Badges
- Science topic
- Similar topics
- Biological Science
- Anatomy
- Tissue