Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $a,b,c$ be positive real numbers. Prove that $a^3+b^3+c^3\geq a^2b+b^2c+c^2a$.

My (strange) proof:

$$ \begin{align*} a^3+b^3+c^3 &\geq a^2b+b^2c+c^2a\\ \sum\limits_{a,b,c} a^3 &\geq \sum\limits_{a,b,c} a^2b\\ \sum\limits_{a,b,c} a^2 &\geq \sum\limits_{a,b,c} ab\\ a^2+b^2+c^2 &\geq ab+bc+ca\\ 2a^2+2b^2+2c^2-2ab-2bc-2ca &\geq 0\\ \left( a-b \right)^2 + \left( b-c \right)^2 + \left( c-a \right)^2 &\geq 0 \end{align*} $$

Which is obviously true.

However, this is not a valid proof, is it? Because I could just as well have divided by $a^2$ rather than $a$:

$$ \begin{align*} \sum\limits_{a,b,c} a^3 &\geq \sum\limits_{a,b,c} a^2b\\ \sum\limits_{a,b,c} a &\geq \sum\limits_{a,b,c} b\\ a+b+c &\geq a+b+c \end{align*} $$

Which is true, but it would imply that equality always holds, which is obviously false. So why can't I just divide in a cycling sum?

Edit: Please don't help me with the original inequality, I'll figure it out.

share|cite|improve this question
You can't assume what you want to prove. – user60887 Jul 7 '13 at 23:26
@user60887 I'm not doing that, I'm trying to reduce it to something that I can prove. – Tim Vermeulen Jul 7 '13 at 23:27
@timvermeulen You cannot divide with $a$ the cyclic sum is an simpler way to write to expressiong $a^3+b^3+c^3$ since you cannot divide with $a$ in this expression you cannot divide in your other expression(with the cyclic sum symbol). Until you feel comfortable with another way of writing the same thing, first translate what an operation means in the expression where you are familiar with. – clark Jul 7 '13 at 23:31
for example write $\sum _{a,b,c} a^2 \geq \sum_ {a,b,c}a, \Rightarrow \sum _{a,b,c} a \geq \sum_ {a,b,c}1$ which is false for $a=b=c=0$ – clark Jul 7 '13 at 23:34
The inequality is obviously true if a=b=c so due to symmetry, why not consider a>b ? That is, write a = b + k with k>0 substitute for a and see if the inequality becomes easier to handle. (It is just a hunch, I am not sure if it works...) – imranfat Jul 7 '13 at 23:41
up vote 6 down vote accepted

Just assume, wlog $a\leq b\leq c$. Then this equation is all you need: $$a^3+b^3+c^3=a^2b+b^2c+c^2a+\underset{\geq 0}{\underbrace{(c^2-a^2)(b-a)}}+\underset{\geq 0}{\underbrace{(c^2-b^2)(c-b)}}\geq a^2b+b^2c+c^2a$$

share|cite|improve this answer
but if $b \le a \le c$, this method doesn't work. – chenbai Jul 9 '13 at 8:30
This is what the wlog is about. As the equation is somehow symmetrical, you can use $$a^3+b^3+c^3=a^2b+b^2c+c^2a+(c^2-b^2)(c-a)+(a^2-b^2)(a-b)$$ in this case. – Tomas Jul 9 '13 at 8:59
OK, that is nice. I simply swap $a$ and $b$ and get wrong result. – chenbai Jul 9 '13 at 9:06
This is because, it is not totally symmetrical in $a^2b+b^2c+c^2a$, you do not get the same expression here by arbitrarily swapping. – Tomas Jul 9 '13 at 9:16

Without making any assumption, just simple $AM\ge GM$ $$a^3+a^3+b^3\ge3a^2b$$ $$b^3+b^3+c^3\ge3b^2c$$ $$c^3+c^3+a^3\ge3c^2a$$ $$a^3+b^3+c^3\ge a^2b+b^2c+c^2a$$

share|cite|improve this answer

WOLG, Let $c$=Max{$a,b,c$}, then there is 2 cases:

case I: $0<a \le b \le c$, we want to prove $c^2(c-a) \ge a^2(b-a)+b^2(c-b)$

we have $c^2\ge b^2, c^2\ge a^2 \to $,RHS $\le c^2(b-a)+c^2(c-b)=c^2(c-a)$

case II: $0<b \le a \le c$, we want to prove $a^2(a-b)+c^2(c-a) \ge b^2(c-b)$

we have $a^2\ge b^2,c^2 \ge b^2, \to$LHS $ \ge b^2(a-b)+b^2(c-a)=b^2(c-b)$

to summary 2 cases, we have $a^2(a-b)+b^2(b-c)+c^2(c-a) \ge 0$


share|cite|improve this answer

Your Answer


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.