home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!mcsun!sunic!uts!biobase!pamaga
- From: pamaga@biobase.aau.dk (Paulo Magalhaes)
- Subject: Best fit to a curve
- Message-ID: <C17n3w.AEr@biobase.aau.dk>
- Organization: The Danish BioBase
- Date: Thu, 21 Jan 1993 15:31:08 GMT
- Lines: 25
-
- Hi,
-
- The subjet line says it all... Just like one can use linear regression
- to find a formula to represent a line based on a given set of points, I
- need to do the same thing, only in my case it's a curve. I'll explain
- better. Given a set of points, I can plot them on good old graph paper
- and obtain a very simple curve; this can then be used to calculate, say,
- an unknown Y for a given X. The problem is, I can't get a machine to do
- this!!! Is there a formula (as in LR) which I can introduce in a simple
- program, so that I can start saving on graph paper ;-)
-
- Better still, is there a program which already does this? I mean, I
- supply a set of (x,y), the program works out the formula for the best
- fit curve, and then I can ask "C'mon slave, gimme the value of 'y' when
- 'x'=..." :-)
-
- Please email your replies, as I don't usually read this newsgroup.
-
- Thanks for the attention
-
- Regards
-
- Paulo
- pamaga@biobase.aau.dk
-
-