R help - Jun 2011 - Model Formulae Evaluation

Hello All,

I have searched but haven't been able to find an answer to this question.
I'm writing a function that needs to be able evaluate the right-hand side of
a general non-linear formula for different parameter values, similar to what
nls() does. I looked through the nls() code but it isn't clear what is going
on. Could someone point me to a reference or a way I might do this?

Thanks

--
View this message in context:
http://r.789695.n4.nabble.com/Model-Formulae-Evaluation-tp3610328p3610328.html
Sent from the R help mailing list archive at Nabble.com.

Please allow me to clarify my original question. What I really need to be
able to do it is to take arbitrary functions and evaluate them for arbitrary
parameter values. I'm doing the optimization myself, so I need to be able to
take a user's function and evaluate them at the current parameter values
during my optimization process. So it would look something like this:

opt.fun <- function(user.formula, param.values)
{
    #--- I would do some optimization here ---#

    fitted.values <- eval.fun(user.formula, param.values) ##<---- this is
what I need
}

Where fitted.values is a vector of the same size as the x-values in
user.formula. nls() does this somehow. I could do this easily myself if I
have the user pass the formula in reverse polish notation, but I was hoping
there was a more canonical was to do this in R.

--
View this message in context:
http://r.789695.n4.nabble.com/Model-Formulae-Evaluation-tp3610328p3611329.html
Sent from the R help mailing list archive at Nabble.com.

On Mon, Jun 20, 2011 at 1:31 PM, albeam <beam.andrew at gmail.com>
wrote:
> Hi everyone,
>
> Thank you for the help, I apologize for not "providing commented,
minimal,
> self-contained, reproducible code." I was looking for some pointers
about
> how to do this in general, but it would have been helpful for me to post a
> specific example. Anyway, after the feedback this is the solution I have
> settled on for future reference:
>
> ## Simple Linear Model Example ##
> X <- c(1,2,3,4,5)
> user.formula <- Y ~ a + b*X
>
> param.values <- list(a=1.5, b=3, x=X)
> eval(user.formula[[3]], envir = param.values)
>
> #Output#
> [1] ?4.5 ?7.5 10.5 13.5 16.5
>
> ## Something a little more complicated ##
> X <- c(.01,.1,1,10,25,50,100)
>
> user.formula <- Y ~ Top - (Top-Bot)/(1+(X/b)^W)
> param.values <- list(Top=100, Bot=1, b=.05, W=2, x=X)
>
> eval(user.formula[[3]], envir = param.values)
>
> #Output#
> [1] ?1.039584 ?4.807692 80.200000 99.753117 99.960416 99.990101 99.997525
>
> Thanks again.
>
Also try this:
> library(gsubfn)
> X <- c(.01,.1,1,10,25,50,100)
> fo <- Y ~ Top - (Top-Bot)/(1+(X/b)^W)
> FUN <- fn$identity(fo[-2]) # drop LHS
> # and then in your inner loop:
> param.values <- list(Top=100, Bot=1, b=.05, W=2, X=X)
> do.call(FUN, param.values)
[1]  4.807692 80.200000 99.753117 99.997525 99.999604 99.999901 99.999975

where we note that we only had to create FUN once.

-- 
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com