Am 02.07.2018 um 16:46 schrieb Riccardo
(Jack) Lucchetti:
On
Mon, 2 Jul 2018, Sven Schreiber wrote:
Am 02.07.2018 um 11:59 schrieb Andreas
Zervas:
I found out that although cointegration
case 3 (unrestricted constant) works well, case 4 (trend
restricted to the cointegration space) produces the following
error message:
Matrices not conformable for operation
*** error in function vecm_est, line 87
> matrix mu = olspar[1,] | (jalpha * jbeta[n+1,])'
Hm, this looks wrong indeed as a matrix product. Do I guess
correctly that you have a rank higher than 1? If so, then it
could be that this case has never been properly tested.
The same problem would probably also happen for a restricted
constant, could you try that just to check (also with rank >
1)?
It would be quite helpful if you could send us a script exposing
the problem.
If my hunch is correct then it should be exposed for any SVEC
specification with r>1 and d==2 or d==4. But I have no time to
test right now.
(Knowing that "you" wasn't me.)
cheers,
sven