estat sbcusum error

[Fabiana De Cristofaro]

Hi all!
I run ardl y x1 x2 x3 x4, ec1 estimating a ARDL(1,0,2,2,1) with 50 obs.
The model is well specified with all post estimation, including cusum6, working.

By when I run estat sbcusum I get this error:

" r(3200) conformability error function returned error "

I use Stata 17 and in some other regressions It worked. I would like to know what does it mean that error, whether it is possible to fix and otherwise if the cusum6 plot without any test is enough to convince a referee that model parameters are stable.

Thank you all in advance

Fabiana

[Daniel Schaefer]

The help file says "estat sbcusum requires that the current estimation results be from regress", so it may be a compatibility issue with ardl.

[Fabiana De Cristofaro]

Thanks a lot Mr. Schaefer, of course I tried but I get the same.
That's why I was thinking about something wrong in my data but I really cannot understand what. Maybe too few degrees of freedom? Too few observations?