MathGroup Archive 2001

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

Search the Archive

Re: non-linear regression

  • To: mathgroup at
  • Subject: [mg30160] Re: non-linear regression
  • From: Jens-Peer Kuska <kuska at>
  • Date: Wed, 1 Aug 2001 02:19:14 -0400 (EDT)
  • Organization: Universitaet Leipzig
  • References: <9k5r0p$hbg$>
  • Sender: owner-wri-mathgroup at




"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"


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

  • Prev by Date: Re: Solving linear equations with symbolic RHS
  • Next by Date: Re: Lists and speed
  • Previous by thread: Re: Solving linear equations with symbolic RHS
  • Next by thread: Re: Lists and speed