Is recursive and rolling forecast possible with VECM (vector error correction model)?

[Raymond James]

Dear Statalists,

I was wondering if recursive and rolling forecasting is possible with VECM? I used the below commands to generate a 3 step forecast but unable to do a recursive or a rolling forecast on VECM using STATA. I am using STATA/SE 16.0 for Windows.

vec $Y if tin(,2009m1), lag(5)
fcast compute M_, step(3) replace
fcast graph M_lnPn M_lnPx , observed lpattern(dash) ///
saving(sample.gph, replace)

I would really appreciate it if someone could guide me the way if there is a method on STATA.

Thanks in advance.