0
$\begingroup$

I would like to forecast iteratively a model with a lag dependent variable and a recursive independent variable. Is this possible to do in Stata?

Basically, it would be:

y(t+1)=a0+a1*y(t)+a2*x(t),

as usual, for 1 step ahead forecast (I know how to do this part), but in the 2nd step it would be

y(t+2)=a0+a1*y(t+1)+a2*x(t+1),

where x(t+1) is a forecast from another model, whereas y(t+1) is the iterative forecast from the previous period

$\endgroup$
  • $\begingroup$ SAS can't do this and since STATA is (generaly) a SAS knockoff I presume they can't BUT I have used other software that actually does this. $\endgroup$ – IrishStat Sep 21 '17 at 19:39
  • $\begingroup$ Such as? Can it be done in R? $\endgroup$ – Roux Sep 22 '17 at 4:38
  • $\begingroup$ Apparently, forecast solve does it by default in cases with a lag dependent and an exogenous variable. Thanks! $\endgroup$ – Roux Sep 22 '17 at 9:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.