I`m trying to make a ARDL model and have 6 variables were the dependent variable are I(0), stationary at level, with 6 lags and the 5 remaining independent variables are I(1), stationary at first differances, with 1 lag. I was reading in a blog (https://nomanarshed.wordpress.com/2014/11/16/a-manual-for-ardl-approach-to-cointegration/) that the dependent variable has to be I(1) in order for the model to behave better. Do anyone know if this is correct? And if it is correct, does it mean I can not use ARDL to model my data? The main reason for me wanting to use ARDL model is since it can handle both I(0) and I(1) variables.

I would really appriciate if any have any tips :) Thanks in advance


1 Answer 1


It does not 'have to', but this will be a degenerate case (see Pesaran, Shin & Smith, 2001, p. 294).

Two degenerate cases arise. First, [. . .] $y_t$ is (trend) stationary or $y_t\sim I(0)$ [. . .]. Consequently the difference variable $\Delta y_t$ depends only on its own lagged level $y_{t-1}$ in the conditional ECM (8) and not on the lagged levels of $\bf{x}_{t-1}$ of the forcing variables. Second, [...]


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.