0
$\begingroup$

Let $[\cdot]$ denote the integral part of a real number, that is the only integer $a$ satisfying

$a \leq x < a+1$.

Let $0<r<1$ be a rational number and $m \geq 0$ a positive integer.

How do you prove that the following equality?

$[rm]+[(1-r)m]=m-1$

This should be completely elementary but I don't manage!!

$\endgroup$
1
  • $\begingroup$ It's corrected now $\endgroup$
    – user84782
    Jul 2, 2013 at 18:01

2 Answers 2

2
$\begingroup$

The claimed formula is incorrect. If $r=1/2$ and $m=2$ then the two integer parts sum to $m$ rather than $m-1$.

$\endgroup$
2
  • $\begingroup$ Oh, that's very interesting. Which is the correct formula then? $\endgroup$
    – user84782
    Jul 2, 2013 at 18:02
  • $\begingroup$ I am not sure what you are aiming at here, but at any rate of either of the numbers $rm$ and $(1-r)m$ is not an integer, then the sum will be $m-1$ as stated. Therefore one needs to add a condition to ensure that one of them is not an integer. For example, if you assume that $r$ is irrational instead of rational, that should do it. $\endgroup$ Jul 2, 2013 at 18:04
1
$\begingroup$

This will not be true always. Let $r=\frac 12, m=4$, then $[rm]+[(1-r)m]=4\neq m-1$. But this gives a hint: if $rm$ is integral, we have $[rm]+[(1-r)m]=rm+m-rm=m$, while if $rm$ is not integral we have $rm-[rm]+\{rm\}$ and $[rm]+[(1-r)m]=[rm]+[m-[rm]-\{rm\}]=[rm]+m-[rm]-1$ So the formula is correct for $rm$ non-integral, and the $-1$ should be removed for $rm$ integral.

$\endgroup$
0

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .