Re: Can we find the function?
- From: Robert Israel <israel@xxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
- Date: Thu, 01 Nov 2007 18:21:11 -0500
David W. Cantrell <DWCantrell@xxxxxxxxxxx> writes:
David W. Cantrell <DWCantrell@xxxxxxxxxxx> wrote:
Robert Israel <israel@xxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:[snip]
evan <evan.e.thomas@xxxxxxxxx> writes:
How about f(x) = 1/x if x != 0,
0 if x = 0
f(f(x)) = x, certainly a polynomial function.
... but also a monomial.
Right. But, as preface to my example:
If f(f(x)) were allowed to be a monomial, I would use simply f(x) = 1/x.
[Note that I'm taking the codomain of f to be R*, the one-point extension
of R, so that we have f(0) = oo and f(oo) = 0.] Then f(f(x)) = x.
--------------------------------
Now for the desired example:
f(x) = 2/(x + 1)^2 - 1 yields f(f(x)) = (x + 1)^4/2 - 1
Note that again I am taking the codomain to be R*. Thus, f(-1) = oo and
f(oo) = -1.
Perhaps I should make a comment for anyone who thinks that using R* is
somehow "cheating". If we want to have a _real_ non-polynomial function
f(x) such that f(f(x)) is a polynomial with more than one term, then just
modify my function above slightly to give
| { -1 if x = -1,
| f(x) = {
| { 2/(x + 1)^2 - 1 otherwise.
Then we still have f(f(x)) = (x + 1)^4/2 - 1.
David
Or somewhat more generally consider
f(x) = b (x-a)^(-n) + a for x <> a,
f(a) = a
where n is a positive integer.
Then f(f(x)) = b^(1-n) (x-a)^(n^2) + a.
Similarly,
f(x) = b |x-a|^sqrt(n) + a
where n is an even positive integer that isn't a square.
Then f(f(x)) = b |b|^sqrt(n) (x-a)^n + a.
--
Robert Israel israel@xxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Department of Mathematics http://www.math.ubc.ca/~israel
University of British Columbia Vancouver, BC, Canada
.
- Follow-Ups:
- Re: Can we find the function?
- From: Robert Israel
- Re: Can we find the function?
- References:
- Re: Can we find the function?
- From: evan
- Re: Can we find the function?
- From: Robert Israel
- Re: Can we find the function?
- From: David W . Cantrell
- Re: Can we find the function?
- From: David W . Cantrell
- Re: Can we find the function?
- Prev by Date: Re: tommy's division...a halting problem
- Next by Date: Re: Can we find the function?
- Previous by thread: Re: Can we find the function?
- Next by thread: Re: Can we find the function?
- Index(es):
Relevant Pages
|