MathGroup Archive 1999

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: StartingParameters

  • To: mathgroup at smc.vnet.net
  • Subject: [mg16514] Re: [mg16447] StartingParameters
  • From: BobHanlon at aol.com
  • Date: Tue, 16 Mar 1999 03:59:57 -0500
  • Sender: owner-wri-mathgroup at wolfram.com

In a message dated 3/13/99 6:49:40 AM, midelfor at uiuc.edu writes:

>I am trying to use the option of giving starting parameters for a
>NonlinearRegress command. I can't figure out the correct syntax for
>this. Can anyone let me know the correct way to list the starting values
>for two parameters in NonlinearRegress (or I think it should be the same
>as NonlinearFit also)
>
>Here is what I am trying:
>NonlinearRegress[datatable, p[t], t ,{kon, koff}, StartingParameters ->
>{{kon, 0.2},{koff,10^8}}]
>
>I would like to have the data in the table (datatable) fit to the model
>( p[t]) with the variables  ( t) and the parameters ( kon and koff)
>staring at 0.2 for kon and 10^8 for koff. It doesn't fit when I try
>without giving the starting parameters. What is the correct way to put
>the starting parameters in the above statement?
>

Katarina,

Needs["Statistics`NonlinearFit`"]

?NonlinearRegress

Also, see the on-line help.

data = {{1.0, 1.0, .126}, {2.0, 1.0, .219},
{1.0, 2.0, .076}, {2.0, 2.0, .126}, {.1, .0, .186}};

Without starting values

NonlinearFit[data,
theta1 theta3 x1 / (1 + theta1 x1 + theta2 x2),
{x1, x2}, {theta1, theta2, theta3}]

With starting values

NonlinearFit[data,
theta1 theta3 x1 / (1 + theta1 x1 + theta2 x2),
{x1, x2}, {{theta1, 3.}, {theta2, 15.}, {theta3, 5/6}}]

Your example would then be

NonlinearRegress[datatable, p[t], t ,{{kon, 0.2}, {koff, 10^8}}]


Bob Hanlon


  • Prev by Date: Re: Select in Math.
  • Next by Date: Re: "Solve[x==Erf[x], x]"
  • Previous by thread: StartingParameters
  • Next by thread: Re: StartingParameters