A function is defined in $\mathbb{R}^2$ \begin{align*} f(x,y) = \frac{4x^2+(y+2)^2}{x^2+y^2+1}\end{align*} Find the range of possible values for the function.

Here we can see that when $x=0$ and $y=-2$ the minimum value of the range is $0$. To find the maximum value of the range I have found the partial derivatives and equate them to zero, then I got $y=\frac{4}{3}$ and $x^2= \frac{-25}{9}$, but $x$ cannot be complex ($x \in \mathbb{R}$). So how do we find the max values of this function? Could anyone please give me a hint, is this the way of finding the range of a multivariable function?

  • 1
    $\begingroup$ Lagrange mulipliers $\endgroup$ – Will Jagy Nov 7 '20 at 2:48
  • $\begingroup$ yes, but what are the constraints for this function? $\endgroup$ – Priya Nov 7 '20 at 3:01
  • $\begingroup$ denominator ... $\endgroup$ – Will Jagy Nov 7 '20 at 3:04
  • $\begingroup$ Actually, Lagrange multipliers are used to minimise some function $f(x_1,...,x_n)$ under constraint $g(x_1,...,x_n)=0$. I don't think that's very relevant here since you have no such constraint. $\endgroup$ – YiFan Nov 7 '20 at 4:20
  • $\begingroup$ Note that$$(a-f)(x^2+y^2+1)=(a-4)x^2+(a-1)\bigg(y-\frac{2}{a-1}\bigg)^2+\frac{a(a-5)}{a-1}$$ $\endgroup$ – mathlove Nov 7 '20 at 6:01

You've already noticed that the minimum of $f$ is $0$.

The $x^2$ term in the numerator can be easily removed by considering $$g(x,y):=f(x,y)-4=\frac{\frac{4}{3}-3(y-\frac{2}{3})^2}{x^2+y^2+1}.$$

It is easy to show that $$\max_{x,y}{g(x,y)}=\max_y{g(0,y)=\max_y{\frac{4y-3y^2}{y^2+1}}}=1.$$

Therefore, the range of $f$ is $[0,5]$. The maximum is obtained when $x=0,y=\frac{1}{2}$.

  • $\begingroup$ Yes your answer is completely right. I am a student, I have little doubt part could you please explain to me a little bit? what is the clarification on "$\max_{x,y}g(x,y) =\max_y g(0,y)$"? Is this the clarification, the function $g_{x,y}$ will be large when its denominator is going to small? Moreover, at this time, $x$ is not in the numerator, so we can just equate $x$ to zero to large the equation. Am I right? (sorry for my english) $\endgroup$ – Priya Nov 7 '20 at 10:10
  • 1
    $\begingroup$ Yes, you are totally right. However, note that equating $x$ to zero to "maximize" $g(x,y)$ come after the implicit fact that we are considering only the region where $g(x,y)$ is positive, that is $0<y<4/3$. $\endgroup$ – Noah Tang Nov 7 '20 at 11:04
  • $\begingroup$ Thank you very much for the explanation. $\endgroup$ – Priya Nov 7 '20 at 11:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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