Help with Johansen & VAR/VECM model - new to this!

[thebritishguy]

Good afternoon,

I am VERY new to eviews and econometrics and still trying to figure it all out so please bare with me while I am trying to get my head around everything!

To start, I am trying to asses the effects of foreign aid on the exchange rate of certain countries - one at a time (not panel). I have Exchange rates and 4 further variables.

I have run unit root tests and have found that all but one of the variables is stationary with 1 lag, the other is stationary at level.

I then ran KPSS testing in an attempt to confirm this and found that all of them are stationary at level except for 2, that are stationary at lag 1.

Can I now move on to testing whether these are cointegrated by using the Johansen test? How does the fact that they are stationary at different lags affect my regression and models?

To confirm, if I found that my variables are cointegrated, I use the VECM?


Thank you and sorry again for not being the brightest spark in the econometrics world (at the moment ;) )

[thebritishguy]

Can I run VECM if my variables are of different orders?