Re: [R] Automatic differentiation (was: Re: D(dnorm...)?)

2006-01-26 Thread Anders Nielsen
I can confirm that AD Model Builder is used at the Danish Institute for Fisheries Research, and by fisheries people in and all around the Pacific. On a few occasions I have solved a likelihood optimization problem in AD Model Builder, and then wrapped the binary in an R-package, with data read

[R] Automatic differentiation (was: Re: D(dnorm...)?)

2006-01-26 Thread Hans Skaug
Dear Alberto, There are fisheries people also in Europe using AD Model Builder (Denmark and England for instance), but you are probably right that it is more widespread in North America. There is also effort going on where people try to make assessment models written in ADMB callable from R. b

[R] Automatic differentiation (was: Re: D(dnorm...)?)

2006-01-26 Thread Alberto Murta
On Thursday 26 January 2006 07:56, [EMAIL PROTECTED] wrote: > While symbolic computation is handy, I actually think a more pressing > addition to R is some kind of automatic differentiation facility, > particularly 'reverse mode' AD, which can be spectacular. There are > free tools available for i