Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
1996
*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 1996

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

Search the Archive

Memory & repeated call of Module[...]

  • To: mathgroup at smc.vnet.net
  • Subject: [mg5590] Memory & repeated call of Module[...]
  • From: buttgereit at t-online.de (Peter Buttgereit)
  • Date: Fri, 27 Dec 1996 01:59:02 -0500
  • Sender: owner-wri-mathgroup at wolfram.com

Dear Mathgroup,
I am working on Confidence Intervals for parameters obtained with "nonlinear 
methods".
One thing I am investigating is resampling: I estimate model parameters from my 
data, bootstrap my initial residuals and combine them with my initial 
"prediction" to a new, "resampled" dataset to be re-evaluated for new parameters 
...

Problem:
Repeated use of NonlinearFit takes a lot of memory - too much.

  How can I make NonlinearFit "forget" the old calculations?

As far as I have understood the disussions in "The book" and those by Todd 
Gayley in the CleanSlate Package I have to clear lokal variables in NonlinearFit 
which are created in its Modules - how can I do that?

Suggestions appreciated,
greetings from Germany,
Peter


  • Prev by Date: MMA 3.0 benchmark, test notebook
  • Next by Date: Evaluation Group
  • Previous by thread: MMA 3.0 benchmark, test notebook
  • Next by thread: Evaluation Group