Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
2005
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*November
*December
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 2005

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

Search the Archive

Re: errors on parameters from NonlinearRegress

  • To: mathgroup at smc.vnet.net
  • Subject: [mg60739] Re: errors on parameters from NonlinearRegress
  • From: Antti Penttilä@smc.vnet.net
  • Date: Tue, 27 Sep 2005 03:45:12 -0400 (EDT)
  • Organization: University of Helsinki
  • References: <dh2u46$dbd$1@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

contact at dantimatter.com wrote:
> Hello all,
> 
> If I have a two-parameter fit, does anyone know a way to get
> independent error bars on the two best-fit parameters produced by
> NonlinearRegress?


The NonlinearRegress reports the ParameterCITable. The Confidence intervals (error bars) are in that table in the column 'CI'. Usually these error bars are not independent because there is correlation between the model parameters, seen e.g. in AsymptoticCorrelationMatrix that the function also prints.

Regards,
Antti


-- 
   Antti Penttilä       Antti.I.Penttila at helsinki.fi.removethis
     Researcher                     Observatory
tel. +358 50 5240968        00014 University of Helsinki


  • Prev by Date: Using the solution of a differential equation and minimizing the solution.
  • Next by Date: Re: WORD document
  • Previous by thread: Re: errors on parameters from NonlinearRegress
  • Next by thread: [ANN] MathLink for AppleScript 1.3 available