2
$\begingroup$

Let $x$ be a positive real. I want to find a pair of analytic functions $s(x),d(x)$ such that $s(d(x)) = x$ and

$ s(\exp(d(x)))$ ~ $ x + 2 $

More presicely I Also want :

$$ \lim_{x \to \infty} s(\exp(d(x))) - x - 2 = 0$$

Polynomials seem to fail as do polynomials of exp or ln.

Maybe try Lambert-W ?

Or do I want THE impossible ?

|cite|improve this question
$\endgroup$
  • 2
    $\begingroup$ Note that $\lim_n a_n-b_n = 0$ does not mean the same thing as $a_n\sim b_n$. $\endgroup$ – Regret Mar 26 '15 at 13:33
  • $\begingroup$ I want both , so i edited with also $\endgroup$ – mick Mar 26 '15 at 13:40
  • 1
    $\begingroup$ Related: mathoverflow.net/questions/4347/… $\endgroup$ – Jack D'Aurizio Mar 26 '15 at 14:45
  • $\begingroup$ @Jack D'Aurizio Im intrested in dynamics and tetration and I welcome your link therefore. However it might be " in the middle , it cannot be very close to a real iteration of exp. In that sense I think the link is not related. $\endgroup$ – mick Mar 26 '15 at 15:04
  • 1
    $\begingroup$ @mick: I do not think so. Here we have $d(x+2)=e^{d(x)}$ hence $d(x+1)$ is the "functional square root of the exponential function" applied to $d(x)$. $\endgroup$ – Jack D'Aurizio Mar 26 '15 at 15:06
0
$\begingroup$

We have: $$ d(x+2)= e^{d(x)} \tag{1}$$ hence assuming $d(0)=1$ we have that $d$ grows pretty fast: $d(2)=e,d(4)=e^e,d(6)=e^{e^e}$.

Now it is time to look at Anix' answer to this MO question.

|cite|improve this answer
$\endgroup$
  • 1
    $\begingroup$ Your correct. IT is tetration. Tommy posted about the same time to tetration forum this fact. We had a bet where the answer would-be first. Hard call :) $\endgroup$ – mick Mar 26 '15 at 16:48
  • 1
    $\begingroup$ Multitasking is dangerous. Well Thats my excuse ;) $\endgroup$ – mick Mar 26 '15 at 16:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.