For what values of $a$ does the function, $f$, contain periodic orbits, where $f$ is given by: $$f(x)=a+x \mod 1.$$ It seems for any rational number $a$ you get periodic orbits although I don't know how to prove that. Does anyone know if you get periodic orbits if $a$ is irrational? I don't really know anything about dynamical systems so help in approaching this problem would be appreciated.
Asked
Active
Viewed 44 times
1 Answers
1
$f^{n}(x) = n a + x \mod 1$, where $f^{n}$ is $f$ iterated $n$ times. This is $x$ if and only if $n a$ is an integer, so $a$ is an integer divided by $n$, i.e. a rational number.
Robert Israel
- 448,999
-
Does that mean that the set: $$\lim_{k\rightarrow\infty}\left{x,f(x),f^2(x),...,f^k(x)\right}$$ is the real number line, [0,1], if $a$ is irrational? – Peanutlex Feb 05 '19 at 20:48
-
1It doesn't mean that, but it's true. – Robert Israel Feb 05 '19 at 21:15
-
How could we show that it's true if you don't mind me asking? – Peanutlex Feb 05 '19 at 21:26
-
1E.g. it follows from the equidistribution theorem. – Robert Israel Feb 05 '19 at 21:48
-
But if: $$\lim_{k\rightarrow\infty}\left{x_0,f(x_0),f^2(x_0),...,f^k(x_0)\right}=[0,1],$$ then wouldn't that make the set $[0,1]$ a countable infinite set so it cannot be true? – Peanutlex Feb 06 '19 at 09:55
-
@Peanutlex I'm taking $\lim$ in the sense that every point of $[0,1]$ is a limit point of members of the sequence $f^k(x_0)$, not that every point of $[0,1]$ is actually in the sequence. – Robert Israel Feb 06 '19 at 13:16