MathGroup Archive 2001

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

Search the Archive

Fitting multiple nonlinear equations to unequal length datasets

  • To: mathgroup at smc.vnet.net
  • Subject: [mg29789] Fitting multiple nonlinear equations to unequal length datasets
  • From: "Dr J. Satherley" <js1 at liverpool.ac.uk>
  • Date: Tue, 10 Jul 2001 20:25:28 -0400 (EDT)
  • Sender: owner-wri-mathgroup at wolfram.com

Dear All

I just been making quite extensive use of the mrfit package from Mathsource.
Its a very nice little resource making it rather easy to do multiple
non-linear equation fits to datasets where there are some common parameters
in the equations. However, it seems to assume that the number of data points
per set of data are all equal, i.e for every line of the data: {{x1,
x2,..,xn,.{y1,y2,...,yn}} in the dataset they all have the same value of n.

I have a problem in which some data points are missing i.e there are some
lines where the number of x and y values are less than n compared to the
other lines in the dataset. Does anyone know:
1) Of a clever way to get mrfit to recognise such a dataset?
2) If there is a symbol that can be added where the data are missing so that
the mrfit function simply ignores the missing points?
3) If I have to resort to writing my own codes with FindMinimum etc. to
overcome my problem?

Cheers
John

*************************************************************
Dr J. Satherley
Department of Chemistry
University of Liverpool
Liverpool L69 7ZD
United Kingdom
tel: +44(0)151 794 3530
fax: +44(0)151 794 3588 (Dept.)
email: js1 at liv.ac.uk
*************************************************************


  • Prev by Date: Re: TeXSave, TeXForm, etc.
  • Next by Date: Re: Partitioning 1-D list into 2-D Matrix
  • Previous by thread: Re: Can This Be Made Cleaner
  • Next by thread: Combining Plots in a Natural Manner