The Johansen approach with a mixture of I(0) and I(1)

[hetero]

Hi,

I would like to examine if 5 variables are cointegrated. The problem is that 4 of them are I(1) (3 independent variables plus the dependent variable) and one is I(0)(independent variable). I know that a cointegration can be established (without worrying about the integration order of the variables) within the framework of the ARDL approach to cointegration, but I want to make use of the Johansen technique.

It is known that the Johansen approach can be applied with a mixture of I(0) and I(1) variables but the likelihood testing procedure for the cointegration rank is sensitive to I(0) variables.

Rahbek and Mosconi (1999) offer a remedy to the above problem, by suggesting the inclusion of the cumulated explanatory regressors in the error correction term

Could you provide me an idea on how can I handle the above situation or how is it possible to apply Rahbek and Mosconi remedy in E-views?

Rahbek, A. and Mosconi, R. (1999). Cointegration Rank Inference with Stationary regressors in VAR Models. Econometrics Journal 2, 76-91.


Kind Regards