Proving Inequalities with n > 2: A Challenge

phonic
Messages
28
Reaction score
0
Dear all,

I want to prove that the following inequalities are true. I hope you can give some hints. Thanks a lot!

Define

<br /> c_{\beta}=\sum_{j=1}^n<br /> \sigma_j^{\frac{2\beta}{\beta+1}}\sum_{1\leq i&lt;k \leq n}\Big(<br /> \sigma_k^{\frac{2}{3(\beta+1)}} +<br /> \sigma_i^{\frac{2}{3(\beta+1)}} \Big)^3 , \mbox{\hspace{1.5cm}}0\leq\sigma_i^2\leq 1, \forall i\in \{1,2,\cdots,n\}<br />


Prove that:

1)
<br /> n \sum_{1\leq i &lt;k \leq n}\Big(<br /> \sigma_k^{\frac{2}{3}} +<br /> \sigma_i^{\frac{2}{3}} \Big)^3 \geq \sum_{j=1}^{n}\sigma_j \sum_{1\leq i &lt;k \leq n}\Big(<br /> \sigma_k^{\frac{1}{3}} +<br /> \sigma_i^{\frac{1}{3}} \Big)^3<br />


and
2)
<br /> 4n(n-1) \sum_{j=1}^n \sigma_j^2 \geq \sum_{j=1}^{n} \sigma_j \sum_{1\leq i &lt;k \leq n}\Big(<br /> \sigma_k^{\frac{1}{3}} +<br /> \sigma_i^{\frac{1}{3}} \Big)^3<br />

I.e.:
<br /> c_0 \geq c_1<br />

<br /> c_{\infty} \geq c_1<br />


It is easy to prove when n=2 by taking the dirivative with respect to \sigma_1, and showing that the dirivative switches the sign at point \sigma_1 = \sigma_2. How to prove when n>2?

Thanks a lot!
 
Physics news on Phys.org
1) So, what happens if you let B=0, then B=1 ? Base on your given condition C0>C1, you can prove 1.
2) Things are just the same, you pay attention to 4n(n-1) which is kind of a sum's result, and let B->infinity, then you will have C_infinity that again, under given condition C_inf > C1. You can prove 2.
 
Last edited:
Thanks for your sugestion. I want to prove that both C_0 and C_inf are larger than C_1. Could you give more detailed hints? Thanks a lot!

Emieno said:
1) So, what happens if you let B=0, then B=1 ? Base on your given condition C0>C1, you can prove 1.
2) Things are just the same, you pay attention to 4n(n-1) which is kind of a sum's result, and let B->infinity, then you will have C_infinity that again, under given condition C_inf > C1. You can prove 2.
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top