[R-SIG-Finance] CONSTRAINED REGRESSIONS
markleeds2 at gmail.com
Sat Aug 15 22:36:05 CEST 2015
Hi: There probably is but I don't know of the package.
Another way is just to maximize the standard likelihood with the
constraints that all of the coefficients are between 0 and 1, except for
one of them which is then 1 minus the sum of the others. Note that,
whatever you use, don't use L-BFGS-B. It's buggy and not due to anything
anyone associated with R did. It uses a Fortran algorithm and Nocedal
posted the bug a few years ago.
As a recommendation, I've had very good experiences with Rvmmin ( use the
variable metric approach just like BFGS ) so I'm biased towards that one.
But, if you want to use it, speak with John Nash because it's often in
active development-improvement so you should get the latest one.
On Sat, Aug 15, 2015 at 3:00 PM, Eric Weigel <weigel.eric1 at gmail.com> wrote:
> I wish to run a regression where the slope coefficients are bound between
> 0 and 1, the sum of the slope coefficients sums to 1 and there is an
> intercept (no bounds). Specifically I wish to run a Heston/Rouwenhorst type
> of regression using country and sector dummies.
> Is there a procedure in R that can run such a constrained regression?
> Thanks for your help.
> *Eric J. Weigel*
> [[alternative HTML version deleted]]
> R-SIG-Finance at r-project.org mailing list
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions
> should go.
[[alternative HTML version deleted]]
More information about the R-SIG-Finance