0
$\begingroup$

It is probably a simple question, but I wasn't able to solve it.

Let $n=10$ and $l>n$ an integer. Let $a>0$ a real number such that $\{a\}>\frac{1}{l}$ (where $\{ \cdot \}$ denotes the fractional part).

I would like to prove (I think this is true) that

There exists an integer $k \in \{1,\dots, l\}$ such that $\{ka\} \in [1/n,2/n]$.

Since $l>n$, I know that at least two numbers from the set $X=\{ \{ka\} \mid k \in \{1,...,l\}\}$ belong to an interval of the form $[r/n,(r+1)/n]$. I don't know how to go further.

Thank you for your help!

$\endgroup$
  • $\begingroup$ This is not true. Try $l>10$ arbitrary and $a=\{a\}=\frac12>\frac1l$. Then $\{ka\}$ is either $0$ or $\frac12$, but never $\in[\frac1{10},\frac2{10}]$. $\endgroup$ – Hagen von Eitzen Feb 27 '16 at 21:04
0
$\begingroup$

This is not true.

Consider $l = 11$. Let $a = 2.5$. Clearly, $0.5 = \{2.5\} > \frac{1}{11}$.

Consider $k \in \{1,..., 11\}$. If $k$ is odd, then $\{2.5k\} = 0.5$. If $k$ is even, then $\{2.5k\} = 0$.

Since $0$ and $0.5$ are not contained in $[\frac{1}{10}, \frac{2}{10}] = [0.1,0.2]$, the statement is false.

$\endgroup$

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.