1
$\begingroup$

For some $w$,$x$,$y$,$z$ it holds that $$w+x>y+z. \quad \quad (1)$$Does that mean that $$\sqrt w+\sqrt x> \sqrt y+\sqrt z \quad \quad (2)$$ and the other way around, so if (2) holds then (1) too?

Thanks!

$\endgroup$
1
  • $\begingroup$ $$\sqrt{25} + \sqrt{25} > \sqrt{49} + \sqrt{4}$$ $\endgroup$ – AlohaSine Jun 14 '16 at 17:07
1
$\begingroup$

No. Counter-examples found by computer search:

$$1 \implies 2 \text{ counter-example } (w, x, y, z) = (1, 6, 3, 3)$$ $$2 \implies 1 \text{ counter-example } (w, x, y, z) = (2, 2, 1, 3)$$

$\endgroup$
0
$\begingroup$

(1) does not imply (2). For example $1 + 9 > 6 + 3$ but $\sqrt{6} + \sqrt{3} > \sqrt{1} + \sqrt{9}$. To find this counterexample I generated 4-tuples of random single-digit integers $(w, x, y, z)$ between 1 and 9 until one was a counterexample.

Given this, there's no need to find a separate counterexample for (2) implies (1). If

$$w + x > y + z \textrm{ and } \sqrt{w} + \sqrt{x} < \sqrt{y} + \sqrt{z}$$,

then, by reversing the inequalities,

$$\sqrt{y} + \sqrt{z} > \sqrt{w} + \sqrt{x} \textrm{ and } y + z < w + x.$$

$\endgroup$

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.