I am currently working on an ARDL model: $$ e_t = \sum(e_t) + \sum(GDP_t) + \sum(t_t) + \epsilon_t $$ where $e$ is electricity demand, and $t$ is degree.

Can someone outline the assumptions that the variables must satify in the ARDL model and tests for cheching these assumptions?

I know that in this model the variables could be $I(0)$ or $I(1)$ but not $I(2)$.

What are the other assumptions and corresponding tests for the variables?


1 Answer 1


Yes, ARDL need no pretesting of order of integration of the variables, and that it is applicable for a mixture of I(0) and I(1) variables with no variable being I(2). But, it requires that the dependent variable should be I(1).

  • 1
    $\begingroup$ In the above, I just want to make a couple of corrections. the dependent variable does not need to be I(1). Also, the order of the LHS and RHS just has to be the same. $\endgroup$
    – mlofton
    Commented Sep 21, 2022 at 11:06

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.