The trick to treat the unit root vaiables

Hi guys,

I am simulating a model which includes some unit root variables which obviously don’t get back to the steady state. What is the trick to enter this sort of equations (such as i= k- k(-1) ) in to Dynare to avoid this problem.

In this source ( it talks about “ONE solution” for this problem but I do not know exactly what it means!

Any idea?!!


People typically detrend the model by dividing everything by the stochastic trend.

Thanks for your reply. That would be great if you could please send me an example article/model for this case.


In the unstable snapshot in the examples folder you will find a New Keynesian model that has been treated this way, including the detailed articles showing how to do it.

Dear jpfeifer,

Thanks for your reply. I think you mean this article , “The econometrics of DSGE models”, I am reading it at the moment to see how it works.