home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.programming:2355 comp.realtime:964
- Newsgroups: comp.programming,comp.realtime
- Path: sparky!uunet!cs.utexas.edu!wupost!darwin.sura.net!Sirius.dfn.de!Urmel.Informatik.RWTH-Aachen.DE!messua!dak
- From: dak@messua.informatik.rwth-aachen.de (David Kastrup)
- Subject: Re: Fitting a gaussian curve to data
- Message-ID: <dak.714052745@messua>
- Keywords: fit,gaussian,normal
- Sender: news@Urmel.Informatik.RWTH-Aachen.DE (Newsfiles Owner)
- Nntp-Posting-Host: messua
- Organization: Rechnerbetrieb Informatik / RWTH Aachen
- References: <1992Aug17.100836.15450@jet.uk>
- Date: 17 Aug 92 11:59:05 GMT
- Lines: 14
-
- rmal@jet.uk (Richard Lucock) writes:
-
- >Hi,
- > I am looking for a routine (code or algorithm) to fit a
- >gausian curve to data. I can do this easily using a least squares
- >minimisation routine (from the Nag library in fact), but it is too
- >slow (this is for a realtime application, hence the crossposting), so
- >I am trying to find something faster.
-
- Have you tried the simple-minded approach of calculating mean
- and variance of your data and using them as parameters to your
- gaussian? BTW, doing this numerically stable is also discussed in
- Seminumerical Algorithms, Vol. 2 of The Art of Computer Programming
- by Knuth.
-