I reworked this using the innovations form and it seems to work. But none of
the parameters were changed after estimation.
summary lists:
$counts
function gradient
1 1
It guess it calculates the function but doesn't optimize the parameters.
I just used: tfn.est <- estMaxLik(
I thought that I wanted the non-innovations form but wanted R and w(t) to be
zero. I
thought leaving out R would give that. What I'm trying to do is estimate a
transfer function
noise model. I noticed that you suggested to another person to use ARMA()
and one can estimate the forecast function of
I think the problem here is that you seem to be trying to specify a
non-innovations form model, for which both Q and R need to be specified,
but you have only specified Q.
The code guesses whether you are specifying an innovations model or an
non-innovations model based on whether you specify
I can't identify the problem. The package author and maintainer,
Paul Gilbert, might be able to help.
Have you tried "debug(SS)"? This will allow you to walk through
the function line by line looking at things, etc. This often produces
enlightenment.
Alternatively, have
I tried to specify a model in dse1 but something isn't right. Anybody
have any tips?
model<-SS(F=f,G=g,H=h,Q=q,z0=z,P0=p)
Error in locateSS(model$R, constants$R, "R", p, p, plist) :
The dimension of something in the SS model structure is bad.
> dim(f)
[1] 5 5
> dim(g)
[1] 5 1
> dim(h)
[1] 1 5
5 matches
Mail list logo
