An interesting question - I(0) I(1) and I(2)
Posted: Sat Mar 28, 2020 2:34 pm
Hi all,
Here is an interesting question.
What do you think?
I am currently conducting time series analysis, n =36, annual data, for the USA and 4 other countries.
For my other countries, because my unit roots are I(0) and I(1), I have been using ARDL, VAR or VECM.
For my USA variables, with unit root testing, show I(0), I(1) and I(2).
This result has not been found in other countries.
If I cannot use ARDL, VAR or VECM methods because of this unit root result, what method can I use?
If there is no appropriate method, can I manipulate the data (ie: log)?
If I do transfer the data, in this case for my country the USA, how this will reflect for the other countries that do not require transformation?
What do we think?
Thanks all :)
Here is an interesting question.
What do you think?
I am currently conducting time series analysis, n =36, annual data, for the USA and 4 other countries.
For my other countries, because my unit roots are I(0) and I(1), I have been using ARDL, VAR or VECM.
For my USA variables, with unit root testing, show I(0), I(1) and I(2).
This result has not been found in other countries.
If I cannot use ARDL, VAR or VECM methods because of this unit root result, what method can I use?
If there is no appropriate method, can I manipulate the data (ie: log)?
If I do transfer the data, in this case for my country the USA, how this will reflect for the other countries that do not require transformation?
What do we think?
Thanks all :)