NIntegrate v.s. NonlinerFit
- To: mathgroup at smc.vnet.net
- Subject: [mg25811] NIntegrate v.s. NonlinerFit
- From: Carmelo La Rosa <clarosa at dipchi.unict.it>
- Date: Sat, 28 Oct 2000 01:41:04 -0400 (EDT)
- Organization: Dipartimento di Scienze Chimiche
- Sender: owner-wri-mathgroup at wolfram.com
Hi,
my problem is the following:
I have difined a function as following, in order to use in a nonlinear
bestfit containing a numerical integration (work on WinNT 4.0 and
Mathematica 3.0)
<<Statistics`NonlinearFit`
pars={a,b,c}(*parameters for the fit*)
pars0={.......}(*starting parameters of pars*)
A(X)=Exp[-a/X]
K(X)=Exp[-b/X]
f[(T_)?Number,(a_)?Number,(b_)?Number,(c_)?Number]:=
(b*K(T)/(K(T)+1)^2)*Exp[-costant*Module[{X},NIntegrate[A(X)*K(X)/(K(X)+1)),{X,Ton,T}]
the output is a error messagge:
NIntegrate::"nlim": "\!\(X$1\) = \!\(T\) is not a valid limit of
integration."
NIntegrate::"nlim": "\!\(X$1\) = \!\(T\) is not a valid limit of
integration."
NIntegrate::"nlim": "\!\(X$1\) = \!\(T\) is not a valid limit of
integration."
General::"stop":
"Further output of \!\(NIntegrate :: \"nlim\"\) will be suppressed
during \
this calculation."
In[19]:=
fitted=NonlinearFit[data,Cp[T,pars],pars,T,MaxIterations->3000,
Errors\[Rule]1.0,InitialGuess\[Rule]pars0,ProgressTrace\[Rule]True]
NonlinearFit::"toomany":
"Too many variables have been provided for the size of the data;
given \!\
\(2\)-tuples for data, given \!\(4\) variables."
Any assistance will be greatly appreciated.
Carmelo la Rosa
--
____________________________________
------------------------------------
Carmelo La Rosa
Department of Chemical Science
Laboratory of Biophysical Chemistry
University of Catania
V.le A. Doria 6, 95125 Catania
ITALY
Tel1. ++39-95-7385114
Tel2. ++39-95-339572
Fax ++39-95-580138
E-Mail clarosa at dipchi.unict.it
------------------------------------
____________________________________