Topper Copies

(a) For $n \geq 2$, let $f: \mathbb{R}^n \to \mathbb{R}$ be given by $f(x_1, x_2, \ldots, x_2) = x_1^2 x_2^2 \ldots x_n^2$ and $c \in \mathbb{R}$. Find the maximum value of $f$ subject to the constraint $g(x_1, x_2, \ldots, x_n) = x_1^2 + x_2^2 + \cdots + x_n^2 - c^2 = 0$. (b) Using (a), prove the inequality $(a_1 a_2 \ldots a_n)^{\frac{1}{n}} \leq \frac{a_1 + a_2 + \cdots + a_n}{n}$ for any positive real numbers $a_1, a_2, \ldots, a_n$.

Aryan Yadav, AIR 31, 2025

a) Using Lagrange method of multipliers, consider

$h(x_1, x_2 .. x_n) = x_1^2 x_2^2 \ldots x_n^2 + \lambda(x_1^2 + x_2^2 \ldots x_n^2 - c^2)$

$\frac{\partial h}{\partial x_1} = 2x_1 x_2^2 \ldots x_n^2 + 2\lambda x_1 = 0$

$\vdots$

$\frac{\partial h}{\partial x_n} = 2x_n x_1^2 x_2^2 \ldots x_{n-1}^2 + 2\lambda x_n = 0$

We get $\lambda = -\frac{x_1^2 x_2^2 \ldots x_n^2}{x_1^2} = -\frac{x_1^2 x_2^2 \ldots x_n^2}{x_2^2} = \ldots = -\frac{x_1^2 x_2^2 \ldots x_n^2}{x_n^2}$

$\Rightarrow x_1^2 = x_2^2 = \ldots = x_n^2$

but $g(x_1, x_2 \ldots x_n) = 0$

$\Rightarrow x_1^2 = x_2^2 = \ldots x_n^2 = \frac{c^2}{n}$

Max of $f = \left(\frac{c^2}{n}\right)^n$

b) We have shown $\Rightarrow$

$x_1^2 x_2^2 \cdots x_n^2 \leq \left(\frac{x_1^2 + x_2^2 \cdots x_n^2}{n}\right)^n$

(Putting value of b & c from above)

Put $a_i = x_i^2$

$\Rightarrow a_1 a_2 \cdots a_n \leq \left(\frac{a_1 + a_2 \cdots a_n}{n}\right)^n$

$\Rightarrow (a_1 a_2 \cdots a_n)^{1/n} \leq \left(\frac{a_1 + a_2 \cdots a_n}{n}\right)$

$a_i = x_i^2 \Rightarrow a_i \geq 0 \in \mathbb{R}$

(AM $\leq$ GM)