Re: " Wanted the r iterated of f(x) = 2*x/(1-x^2) , f^[r](x) "
- From: rusin@xxxxxxxxxxxxxxxxxxxxx (Dave Rusin)
- Date: 7 Oct 2005 14:26:40 GMT
In article <1128665817.093842.197060@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
<alainverghote@xxxxxxxx> wrote:
>Do some of you know a closed form
>for the rth iterate of 2*x/(1-x^2).
.... which equals tan(2 u) if x = tan(u). That is, if f(x)=2x/(1-x^2)
then f(tan(u)) = tan( 2u ), so that f(f(tan(u))) = tan(4u) and
f(f(f(tan(u)))) = tan(8u), etc. In general you can then prove by
induction that f^r ( tan(u) ) = tan( 2^r u ) for positive integers r.
You can write this as f^r( x ) = tan( 2^r arctan(x) ) I suppose but
this is only because, for a given x, the possible values of u all
differ by integer multiples of pi and thus the possible values
of 2^r arctan(x) will also differ by integer multiples of pi, giving
them all the same values in tan() . I say this just to emphasize,
as always must be done, that this analysis is predicated on the
assumption that r is an integer; there is nothing in the defnition
of f alone that tells us what f^r even _means_ when r is
not an integer.
dave
.
- Follow-Ups:
- Re: " Wanted the r iterated of f(x) = 2*x/(1-x^2) , f^[r](x) "
- From: alainverghote
- Re: " Wanted the r iterated of f(x) = 2*x/(1-x^2) , f^[r](x) "
- References:
- " Wanted the r iterated of f(x) = 2*x/(1-x^2) , f^[r](x) "
- From: alainverghote
- " Wanted the r iterated of f(x) = 2*x/(1-x^2) , f^[r](x) "
- Prev by Date: Re: infinity
- Next by Date: Re: infinity
- Previous by thread: " Wanted the r iterated of f(x) = 2*x/(1-x^2) , f^[r](x) "
- Next by thread: Re: " Wanted the r iterated of f(x) = 2*x/(1-x^2) , f^[r](x) "
- Index(es):
Relevant Pages
|