encerg
Sent: Friday, May 15, 2009 2:22 PM
To: avraham.ad...@guycarp.com
Cc: r-help@r-project.org; Douglas Bates
Subject: Re: [R] Optimization algorithm to be applied to S4 classes -
specifically sparse matrices
I suggest you try to translate your constraints into an unconstrained
constrained prob
I suggest you try to translate your constraints into an
unconstrained constrained problem using logarithms, then do "nonlinear
mixed effects modeling" as described in chapters 6-8 of Pinheiro and
Bates (2000). To do this, I would first start with the simpler linear
estimation problem to g
Thank you both very much for your replies. What makes this a little less
straightforward, at least to me, is that there needs to be constraints on
the solved parameters. They most certainly need to be positive and there
may be an upper limit as well. The true best linear fit would have negative
en
Dear Doug, et al.:
What would you recommend for analyzing a longitudinal abundance
survey of 22 species, when the species were not selected at random? A
prominent scientist tried to tell me that mixed-effects modeling is
inappropriate in that case because the species were selected
pur
Dear Avraham:
For problems with many parameters to estimate, I highly recommend
Pinheiro and Bates (2000) Mixed-Effects Models in S and S-Plus
(Springer). This book includes numerous examples showing how to use the
"nlme" package. The value of this book is greatly enhanced by the
On Wed, May 13, 2009 at 5:21 PM, wrote:
>
> Hello.
>
> I am trying to optimize a set of parameters using /optim/ in which the
> actual function to be minimized contains matrix multiplication and is of
> the form:
>
> SUM ((A%*%X - B)^2)
>
> where A is a matrix and X and B are vectors, with X as p
Have you considered the following:
solve(qr(A), B)
I have not tried this with a small toy example, and the "qr"
documentation in the Matrix package seems to suggest it. This solves
the optimization problem you mentioned, as noted in
"http://en.wikipedia.org/wiki/Linea
Hello.
I am trying to optimize a set of parameters using /optim/ in which the
actual function to be minimized contains matrix multiplication and is of
the form:
SUM ((A%*%X - B)^2)
where A is a matrix and X and B are vectors, with X as parameter vector.
This has worked well so far. Recently, I
8 matches
Mail list logo
