home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!paladin.american.edu!darwin.sura.net!aurora.LaTech.edu!ramos
- From: ramos@engr.LaTech.edu (Alex Ramos)
- Newsgroups: sci.math.num-analysis
- Subject: ?: Differentiation and Newton's method on complex-valued functions
- Date: 22 Jan 1993 22:34:48 GMT
- Organization: Louisiana Tech University
- Lines: 25
- Distribution: inet
- Message-ID: <1jpsq8INNbc@aurora.engr.LaTech.edu>
- NNTP-Posting-Host: nebula.engr.latech.edu
- X-Newsreader: TIN [version 1.1 PL8]
-
-
- Suppose I have an arbitrary function H(s), which can be assumed as
- being of the form F(s)/G(s), where F and G are polynomials that very
- likely have complex roots. However, the function is not available in
- explicit form, but it can be numerically evaluated at any point with
- calculator's accuracy (12 digits).
-
- I would like to approximate the roots of F(s) and G(s) so as to express
- H(s) explicitly in terms of s.
- I suppose Newton's method would work satisfactorily,
- unfortunately I don't have a clue on how to approximate the derivative
- of a complex-valued function (note that s is complex as well).
-
- Are there any commonly known algorithms for that?
-
- Thanks in advance for any help. (post or e-mail either is fine)
-
- --
-
- /* ramos@engr.Latech.edu
-
- Fala-se Portugues, tche!
- English spoken here.
- */
-
-