Re: non-linear regression
- To: mathgroup at smc.vnet.net
- Subject: [mg30160] Re: non-linear regression
- From: Jens-Peer Kuska <kuska at informatik.uni-leipzig.de>
- Date: Wed, 1 Aug 2001 02:19:14 -0400 (EDT)
- Organization: Universitaet Leipzig
- References: <9k5r0p$hbg$1@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
Hi, Needs["Statistics`NonlinearFit`"] ?NonlinearRegress "NonlinearRegress[data, model, vars, params] searches for a least-squares fit \ to a list of data according to the model containing the variables vars and \ the parameters params. Parameters may be expressed as a list of symbols or a \ list of lists. When expressed as a list, a parameter may be specified with \ starting value(s) and bounds in one of several different ways: {symbol, \ start} or {symbol, min, max} or {symbol, start, min, max}, where start can be \ single value or a list of two values. The data can have the form {{x1, y1, \ ..., f1}, {x2, y2, ..., f2}, ...}, where the number of coordinates x, y, ... \ is equal to the number of variables in the list vars. The data can also be \ of the form {f1, f2, ...}, with a single coordinate assumed to take values 1, \ 2, .... The Method option specifies the LevenbergMarquardt (default), \ Gradient (steepest descent), Newton, QuasiNewton or Automatic search methods. \ The Automatic method does linear fitting for linear models and \ LevenbergMarquardt nonlinear fitting for nonlinear models" Regards Jens Derek Stoll wrote: > > I am looking to do non-linear regression on an equation with two > variables. > > It is of the form V1 * tanh(v2*L/f) *tanh(v3 * L/f) > > Can anyone point me in the right direction for a good routine? Thank > you, > Derek > dcstoll at us.ibm.com