home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!darwin.sura.net!wupost!zaphod.mps.ohio-state.edu!samsung!balrog!ctron.com!wilson
- From: wilson@ctron.com (David Wilson)
- Newsgroups: sci.math
- Subject: Re: Iteration problem
- Message-ID: <4578@balrog.ctron.com>
- Date: 24 Jul 92 13:30:17 GMT
- Sender: usenet@balrog.ctron.com
- Lines: 38
- Nntp-Posting-Host: web
- To: a_rubin@dsg4.dse.beckman.com (Arthur Rubin)
-
-
-
-
- Thanks for the reply on the iteration problem. Not being well versed in
- the hyperbolic functions, I came up with the solution
-
- (1) h (x) = ((x+sqrt(x^2-4)/2)^p + (x-sqrt(x^2-4)/2)^p).
- p
-
- which, I believe, is equivalent to your more concise
-
- (2) h (x) = 2 cosh(p arccosh(x/2)).
- p
-
- It turns out that for integer p, hp is a polynomial. A few interesting
- properties of hp can be deduced from (1) and (2). For instance, the hp
- are compositionally commutative on the range x >= 2, since
-
- h (h (x)) = 2 cosh(p arccosh((2 cosh (q arccosh(x/2)))/2))
- p q
- = 2 cosh(pq arccosh(x/2))
-
- = h (x)
- pq
-
- Also, (1) is reminiscent of the closed form expression of the Fibonnaci
- numbers, suggesting a similar recurrence, and indeed we find
-
- h (x) = x*h (x-1) - h (x-2)
- p p p
-
- Anyway, I thought it was a cute problem, but I probably wouldn't have
- posted it had I known it had such a straightforward solution.
-
- David W. Wilson (wilson@ctron.com)
-
- Disclaimer: "Truth is just truth...You can't have opinions about truth."
- - Peter Schikele, introduction to P.D.Q. Bach's oratorio "The Seasonings."
-