Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Is it true that for $p\in (1,2)$ the following inequalities holds: $$ 2^{p-1} (|x|^p+|y|^p)\leq |x+y|^p+|x-y|^p \leq 2 (|x|^p+|y|^p)$$ for $x, y \in \mathbb{R}$ ?

Thanks.

share|improve this question

1 Answer 1

up vote 3 down vote accepted

We first show that $x_1^p+x_2^p\geq (x_1^2+x_2^2)^{p/2}$ for non-negative $x_1,x_2$. By homogeneity it's enough to show that $f(t):=x^p+1-(x^2+1)^{p/2}$ is non-negative. Its derivative is $f'(x)=px^{p-1}-\frac p22x(x^2+1)^{p/2-1}\geq px^{p-1}-pxx^{p-2}=0$ since the map $s\mapsto s^{p/2-1}$ is decreasing ($p<2$). Now we apply it to $x_1=\frac{x+y}2$ and $x_2=\frac{x-y}2$. We get $$\left|\frac{x+y}2\right|^p+\left|\frac{x-y}2\right|^p\geq\left(\left|\frac{x+y}2\right|^2+\left|\frac{x-y}2\right|^2\right)^{p/2}=\left(\frac{x^2}2+\frac{y^2}2\right)^{p/2}$$ and since the map $t\mapsto t^{p/2}$ is concave, we get the first inequality.

To get the second one, put $u=x+y$ and $v=x-y$ and apply the first inequality to $u$ and $v$. Since $x=\frac{u+v}2$ and $y=\frac{u-v}2$ we have $$2^{p-1}\left(\left|\frac{u+v}2\right|^p+\left|\frac{u-v}2\right|^p\right)\leq |u|^p+|v|^p$$ so $2^{p-1}(|x|^p+|y|^p)\leq 2^p(|x+y|+|x-y|)$ and we are done.

For $p\geq 2$, the first inequality is reversed.

share|improve this answer
    
If I understood correctly, to get the second inequality, we use the first inequality $2^{p-1} [|u|^p+|v|^p] \leq |u+v|^p+|u-v|^p$ with $u=(x+y)/2$, $v=(x-y)/2$. –  Richard Mar 9 '12 at 18:25
    
Yes, we have to use this substitution. –  Davide Giraudo Mar 9 '12 at 18:26
    
Thanks a lot for answer. –  Richard Mar 9 '12 at 18:28
    
You're welcome. –  Davide Giraudo Mar 9 '12 at 18:29

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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