I know of no existing functions in R that support fitting a
multivariate autoregression while fixing some of the parameters.
Of course, as Simon Blomberg famously said in April 2005, "This is R.
There is no if. Only how." [With library(fortunes), try
'fortune("This
is R")'.] If I had to do fit a multivariate AR today with some
parameters fixed, I might write a function to compute the determinant of
the sample covariance matrix, and give it to "optim" or
"nlminb".
I hope someone else will provide us with an easier way.
hope this helps,
spencer graves
Daniel Medina wrote:
> Dear Colleague,
>
> I would like to set a few AR coefficients (not order) to zero in the
> multivariate AR function (mAr.est; mAr library); however, the manual for
> this function does not provide this information. I would appreciate any
> suggestions along this line.
>
> Thankfully yours,
>
> Daniel C Medina
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html