MHB Can the Root Function Solve Inequalities?

AI Thread Summary
The discussion centers on proving that for a quadratic function f(x) = ax^2 + bx + c, where a, b, and c are positive real numbers, the inequality involving the 1024th root of the product of function values at non-negative real numbers holds true. Participants acknowledge errors in previously suggested solutions, with one user thanking another for pointing out these mistakes. The conversation emphasizes the importance of clarity and accuracy in mathematical proofs. The thread highlights the collaborative nature of problem-solving in mathematical discussions. Overall, the focus remains on the validity of the inequality and the need for precise solutions.
lfdahl
Gold Member
MHB
Messages
747
Reaction score
0
Suppose, that $f(x)=ax^2+bx+c$, where $a$,$b$ and $c$ are positive real numbers. Show, that for all non-negative real numbers $x_1,x_2,…,x_{1024}$

\[\sqrt[1024]{f(x_1)\cdot f(x_2)\cdot \cdot \cdot f(x_{1024})} \geq f\left ( \sqrt[1024]{x_1\cdot x_2\cdot \cdot \cdot x_{1024}} \right )\]
 
Mathematics news on Phys.org
Hint:

It might be of use first to show, that
\[f(x)\cdot f(y)\geq \left (f\left ( \sqrt{xy} \right ) \right )^2\]
 
Suggested solution:

Let us show, that for all $x,y > 0$

\[f(x)\cdot f(y)\geq \left (f\left ( \sqrt{xy} \right ) \right )^2 \;\;\; (1)\]

Indeed,

\[f(x)\cdot f(y) - \left (f\left ( \sqrt{xy} \right ) \right )^2 = ab(x^2y+xy^2-2xy\sqrt{xy})+ac(x^2+y^2-2xy)+bc(x+y-2\sqrt{xy}) \\\\ =abxy(\sqrt{x}-\sqrt{y})^2+ac(x-y)^2+bc(\sqrt{x}-\sqrt{y})^2 \geq 0.\]

Now by $(1)$

\[f(x_1)f(x_2)...f(x_{1024})\geq \prod_{i=1,3,5,…}^{1023}\left ( f\left ( \sqrt{x_i\; x_{i+1}} \right ) \right )^2 \geq \prod_{i=1,5,9,…}^{1021}\left ( f\left ( \sqrt[4]{x_i\;x_{i+1}\; x_{i+2}\; x_{i+3}} \right ) \right )^4 \geq ...\geq \left ( f \left ( \sqrt[1024]{x_1x_2...x_{1024}} \right ) \right )^{1024}\]

The required inequality readily follows.
 
Last edited:
lfdahl said:
Suggested solution:

Let us show, that for all $x,y > 0$

\[f(x)\cdot f(y)\geq \left (f\left ( \sqrt{xy} \right ) \right )^2 \;\;\; (1)\]

Indeed,

\[f(x)\cdot f(y) - \left (f\left ( \sqrt{xy} \right ) \right )^2 = ab(x^2y+xy^2-2xy\sqrt{xy})+ac(x^2+y^2-2xy)+bc(x+y-2\sqrt{xy}) \\\\ =abxy(\sqrt{x}-\sqrt{y})^2+ac(x-y)^2+bc(\sqrt{x}-\sqrt{y})^2 \geq 0.\]

Now by $(1)$

\[f(x_1)f(x_2)...f(x_{1024})\geq \prod_{i=1}^{1023}\left ( f\left ( \sqrt{x_i\; x_{i+1}} \right ) \right )^2 \geq \prod_{i=1}^{1021}\left ( f\left ( \sqrt{x_i\;x_{i+1}\; x_{i+2}\; x_{i+3}} \right ) \right )^2 \geq ...\geq f\left ( \left ( \sqrt{x_1x_2...x_{1024}} \right ) \right )^{1024}\]

The required inequality readily follows.
in my humble opinion in the 1st product i change by 2 in the 2nd by 4 so on (this is not depicted in expression) and the 1st power is 2 in second 4 so on
 
kaliprasad said:
in my humble opinion in the 1st product i change by 2 in the 2nd by 4 so on (this is not depicted in expression) and the 1st power is 2 in second 4 so on

Thankyou, kaliprasad, for pointing out the errors in the suggested solution.(Handshake)
I´m sorry for having posted a solution with severe typos.(Sadface)
 
Seemingly by some mathematical coincidence, a hexagon of sides 2,2,7,7, 11, and 11 can be inscribed in a circle of radius 7. The other day I saw a math problem on line, which they said came from a Polish Olympiad, where you compute the length x of the 3rd side which is the same as the radius, so that the sides of length 2,x, and 11 are inscribed on the arc of a semi-circle. The law of cosines applied twice gives the answer for x of exactly 7, but the arithmetic is so complex that the...
Just chatting with my son about Maths and he casually mentioned that 0 would be the midpoint of the number line from -inf to +inf. I wondered whether it wouldn’t be more accurate to say there is no single midpoint. Couldn’t you make an argument that any real number is exactly halfway between -inf and +inf?

Similar threads

Back
Top