Re: Re: Evaluate and HoldAll
- To: mathgroup at smc.vnet.net
- Subject: [mg24245] Re: [mg24222] Re: Evaluate and HoldAll
- From: Andre Heinemann <andre at aflo4.ifw-dresden.de>
- Date: Tue, 4 Jul 2000 15:22:08 -0400 (EDT)
- Organization: IFW
- Sender: owner-wri-mathgroup at wolfram.com
Dear Andrzej Kozlowski and Carl K. Woll
Thanks for the explanation of my error in reasoning.
The
t[q_,lseq__] := Body[...gr[..,lseq]
way works well for some of my problems.
Andrzej, you were right with your guess of the "why" I want to do this
and this..
I don't want to to bore you to death with all the details of the problem
and so
I extract the "essentials". But I learned, there can't be a common sense
of the
word "essentials" in case of a error in reasoning.
So I will present the problem (because I still have one) today in
detail.
My purpose is to write a Funktion nnregress[]
The aim of this function should be a "simple" and "automatic"
Nonlinear Regression Analysis for NOT analytical functions.
So first I take some special case:
I have a function g[r,a[i]] , were the a[i]'s are the parameters for
fitting
and I have a functional : h[q,g[r,a[i]] ] := Integrate[
...*g[r,a[i]],{r,0,inf]
in general I don't know the number of a[i]'s
so I can define (after your respons)
gr[r,a1,a1,a3] := a1*Exp[a2 - a3 x] -> to expr in Module
param = {{a1, {2., 2.2}}, {a2, {0.1, 0.12}}, {a3, {0.5, 0.52}}} -> to
parameter in Module nnregress
nnregress[....,expr_,parameter_List,...] := Module[{