With $zin mathbb C$ find the maximum value for $lvert zrvert$ such that $lvert z+frac{1}{z}rvert=1$












1












$begingroup$



With $zin mathbb C$ find the maximum value for |z| such that
$$leftlvert z+frac{1}{z}rightrvert=1.$$
Source: List of problems for math-contest training.




My attempt: it is easy to see that the given condition is equivalent
$$lvert z^2+1rvert=lvert zrvert$$
and if $z=a+bi$,
begin{align*}
lvert zrvert&=lvert a^2-b^2+1+2ab irvert=sqrt{(a^2-b^2+1)^2+4 a^2b^2}\
&=sqrt{(a^2-b^2)^2+2(a^2-b^2)+1+4a^2b^2}\
&=sqrt{(a^2+b^2)^2+2(a^2-b^2)+1}\
&=sqrt{lvert zrvert^4+2(a^2-b^2)+1}
end{align*}

I think it is not leading to something useful... the approach I followed is probably not useful.



Hints and answers are welcomed.










share|cite|improve this question











$endgroup$








  • 1




    $begingroup$
    Apologies for the unnecessary edit
    $endgroup$
    – Shubham Johri
    Dec 21 '18 at 20:55
















1












$begingroup$



With $zin mathbb C$ find the maximum value for |z| such that
$$leftlvert z+frac{1}{z}rightrvert=1.$$
Source: List of problems for math-contest training.




My attempt: it is easy to see that the given condition is equivalent
$$lvert z^2+1rvert=lvert zrvert$$
and if $z=a+bi$,
begin{align*}
lvert zrvert&=lvert a^2-b^2+1+2ab irvert=sqrt{(a^2-b^2+1)^2+4 a^2b^2}\
&=sqrt{(a^2-b^2)^2+2(a^2-b^2)+1+4a^2b^2}\
&=sqrt{(a^2+b^2)^2+2(a^2-b^2)+1}\
&=sqrt{lvert zrvert^4+2(a^2-b^2)+1}
end{align*}

I think it is not leading to something useful... the approach I followed is probably not useful.



Hints and answers are welcomed.










share|cite|improve this question











$endgroup$








  • 1




    $begingroup$
    Apologies for the unnecessary edit
    $endgroup$
    – Shubham Johri
    Dec 21 '18 at 20:55














1












1








1


1



$begingroup$



With $zin mathbb C$ find the maximum value for |z| such that
$$leftlvert z+frac{1}{z}rightrvert=1.$$
Source: List of problems for math-contest training.




My attempt: it is easy to see that the given condition is equivalent
$$lvert z^2+1rvert=lvert zrvert$$
and if $z=a+bi$,
begin{align*}
lvert zrvert&=lvert a^2-b^2+1+2ab irvert=sqrt{(a^2-b^2+1)^2+4 a^2b^2}\
&=sqrt{(a^2-b^2)^2+2(a^2-b^2)+1+4a^2b^2}\
&=sqrt{(a^2+b^2)^2+2(a^2-b^2)+1}\
&=sqrt{lvert zrvert^4+2(a^2-b^2)+1}
end{align*}

I think it is not leading to something useful... the approach I followed is probably not useful.



Hints and answers are welcomed.










share|cite|improve this question











$endgroup$





With $zin mathbb C$ find the maximum value for |z| such that
$$leftlvert z+frac{1}{z}rightrvert=1.$$
Source: List of problems for math-contest training.




My attempt: it is easy to see that the given condition is equivalent
$$lvert z^2+1rvert=lvert zrvert$$
and if $z=a+bi$,
begin{align*}
lvert zrvert&=lvert a^2-b^2+1+2ab irvert=sqrt{(a^2-b^2+1)^2+4 a^2b^2}\
&=sqrt{(a^2-b^2)^2+2(a^2-b^2)+1+4a^2b^2}\
&=sqrt{(a^2+b^2)^2+2(a^2-b^2)+1}\
&=sqrt{lvert zrvert^4+2(a^2-b^2)+1}
end{align*}

I think it is not leading to something useful... the approach I followed is probably not useful.



Hints and answers are welcomed.







algebra-precalculus complex-numbers contest-math






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Dec 23 '18 at 18:59









user10354138

7,4322925




7,4322925










asked Dec 21 '18 at 20:39









bluemasterbluemaster

1,653519




1,653519








  • 1




    $begingroup$
    Apologies for the unnecessary edit
    $endgroup$
    – Shubham Johri
    Dec 21 '18 at 20:55














  • 1




    $begingroup$
    Apologies for the unnecessary edit
    $endgroup$
    – Shubham Johri
    Dec 21 '18 at 20:55








1




1




$begingroup$
Apologies for the unnecessary edit
$endgroup$
– Shubham Johri
Dec 21 '18 at 20:55




$begingroup$
Apologies for the unnecessary edit
$endgroup$
– Shubham Johri
Dec 21 '18 at 20:55










6 Answers
6






active

oldest

votes


















1












$begingroup$

Since $$|z| = |z^2-(-1)|geq |z|^2-|-1|$$



Write $r=|z|$ and we have $$r^2-r-1leq 0implies rin [0,r_2]$$ where $$r_{1,2} ={1pm sqrt{5}over 2} $$



So $rleq displaystyle{1+ sqrt{5}over 2}$






share|cite|improve this answer









$endgroup$





















    1












    $begingroup$

    The reverse triangle inequality implies that $|z^2+1|geq ||z^2|-1|$ so that if $|z|^2geq 1$ we see that
    $$|z|geq |z|^2-1$$
    By solving this inequality using the quadratic formula, we see that $|z|leq frac{1}{2}(1+sqrt{5})$, so the maximum must be less than $frac{1}{2}(1+sqrt{5})$. We see that this maximum is attained by setting $z=frac{1}{2}(1+sqrt{5})$.






    share|cite|improve this answer









    $endgroup$













    • $begingroup$
      If $z = frac{1}{2}(1 + sqrt{5})$ then $z^{-1} = frac{1}{2}(-1 + sqrt{5})$ and $|z + z^{-1}| = sqrt{5} ne 1$. The maximum is attained when $z = pm frac{1}{2}(1 + sqrt{5}) i$.
      $endgroup$
      – Daniel Schepler
      Dec 21 '18 at 22:08





















    1












    $begingroup$

    Here's a solution using the dual problem. Assume that the solution of the problem occurs at $z = re^{itheta}.$ Consider
    $$
    F_r(t) = |re^{it}+r^{-1}e^{-it}|^2 = r^2 + r^{-2} + e^{2it}+e^{-2it} = r^2+r^{-2}+2cos 2t.
    $$
    If $z= re^{itheta}$ is the maximizer, we claim that $F_r'(theta)=0$. That is, given $|z|=r$, the angle $theta$ should be chosen so that $F_r(t)$ is either maximized or minimized. To see this, assume to the contrary that $F_r'(theta)neq 0$. Then, there is small $u$ (we allow $u$ to have negative sign) such that $$F_r(theta-u)<1=F(theta)<F_r(theta+u).$$ If it is the case that $rleq 1$, then by choosing slightly large $r'$, we can make $F_{r'}(theta+u)=1$. This contradicts maximality of $r$. If $r>1$, then by choosing slightly large $r'$, we can make $F_{r'}(theta-u) = 1$. This also leads to a contradiction. Now it follows that $cos(2theta)$ should be either $-1$ or $1$. However, we can easily see that $cos(2theta)$ cannot be $1$. Thus, from $cos(2theta)=-1$, we have
    $$
    1=F_r(theta)=r^2 +r^{-2} -2 = (r-r^{-1})^2,
    $$
    or
    $$
    r-1/r = pm 1.
    $$
    This gives
    $$
    r= frac{pm 1+ sqrt{5}}{2}.
    $$
    Therefore, the larger one $r=frac{ 1+ sqrt{5}}{2}$ is the solution.






    share|cite|improve this answer









    $endgroup$





















      1












      $begingroup$

      Suppose $z + frac{1}{z} = e^{it}$ for some $t in mathbb{R}$. Then solving for $z$ gives
      $$z = frac{e^{it} pm sqrt{e^{2it} - 4}}{2} = e^{it} cdot frac{1 pm sqrt{1 - 4 e^{-2it}}}{2}.$$
      Therefore,
      $$ |z| = frac{1}{2} left| 1 pm sqrt{1 - 4 e^{-2it}} right| le frac{1}{2} left( |1| + |sqrt{1 - 4 e^{-2it}}|right) = frac{1}{2} left( 1 + sqrt{1 - 4 e^{-2it}} right)le \
      frac{1}{2} left( 1 + sqrt{|1| + |4 e^{-2it}|}right) = frac{1 + sqrt{5}}{2}.$$

      On the other hand, if we set $t := frac{pi}{2}$ so that $e^{-2it} = -1$ then we will get equality. Therefore, the maximum value of $|z|$ is $frac{1+sqrt{5}}{2}$, which is achieved for $z = frac{1+sqrt{5}}{2} i$. (In fact, a slight refinement of the argument will show that equality holds, i.e. the maximum is achieved, if and only if $z = pm frac{1+sqrt{5}}{2} i$.)






      share|cite|improve this answer











      $endgroup$













      • $begingroup$
        It seems an interesting approach, can you expand a little bit more... how this can lead to the maximum $|z|$.
        $endgroup$
        – bluemaster
        Dec 21 '18 at 21:26






      • 1




        $begingroup$
        OK, I've filled in how the rest of the argument goes.
        $endgroup$
        – Daniel Schepler
        Dec 21 '18 at 22:07



















      1












      $begingroup$

      $|z| =|z+1/z-1/z| le$



      $ |z+1/z| +|1/z| =1+|1/z|;$



      $r:=|z|$;



      $r-1/r le 1;$



      $r^2-r -1le 0;$



      $(r-1/2)^2 -1/4-1le 0;$



      $r-1/2 le (1/2)√5;$



      $r le (1/2)(1+√5)$.



      Is the maximum attained?






      share|cite|improve this answer











      $endgroup$













      • $begingroup$
        Very clear development. What is a simple argument to show that this is not just an upper bound but the maximum?
        $endgroup$
        – bluemaster
        Dec 21 '18 at 22:59










      • $begingroup$
        @RoryDaulton: I'm thinking on a math-contest situation, how could we find an appropriate value that leads to this maximum value. It is not obvious. The solution using the optimization approach appears to me that address this difficulty.
        $endgroup$
        – bluemaster
        Dec 22 '18 at 1:09






      • 1




        $begingroup$
        bluemaster.Easy argument !? I try a bit so easy argument.Triangle inequality in the complex plane: Vector a=z , b=1/z .For side lengths of triangle formed by a,b and a+b (vectors): Since a triangle triangle inequality is strict((>).(lengths of sum of 2 sides of triangle is >than 3rd side.) Equality can only happen if a and b are collinear, We can try z= 1/2(1+√5)(real), does not work, but z= (1/2)(1+√5)i (purely imaginary), works to give |z+1/z|=1.Your thoughts?
        $endgroup$
        – Peter Szilas
        Dec 22 '18 at 7:32










      • $begingroup$
        Very good argument. I was struggling to see by the geometric interpretation and your thoughts clarified. The key is the situation in which a and b are collinear. Thanks!
        $endgroup$
        – bluemaster
        Dec 22 '18 at 9:03










      • $begingroup$
        bluemaster.Thanks! I was struggling too:))Greetings.
        $endgroup$
        – Peter Szilas
        Dec 22 '18 at 9:14



















      0












      $begingroup$

      We can maximize $g(a,b) = a^2+b^2$ subject to $(a^2+b^2)^2 =3b^2-1$. Solve for $a^2$ we have: $a^2= sqrt{3b^2-1} -b^2implies g(a,b) = sqrt{3b^2-1}$. Thus the problem now is the find the maximum value of $b^2$. To this end, let $c = b^2 ge 0$, we have: $a^4+2a^2c+c^2 - 3c+ 1 = 0implies c^2-3c+1 le 0implies c le dfrac{3+sqrt{5}}{2}implies 3b^2 -1 = 3c-1le dfrac{7+3sqrt{5}}{2}implies |z|_{text{max}} = sqrt{g(a,b)}_{text{max}} = sqrt[4]{dfrac{7+3sqrt{5}}{2}}= dfrac{1+sqrt{5}}{2}$,and this max occurs when $a = 0$.






      share|cite|improve this answer











      $endgroup$













      • $begingroup$
        And just to compare with other solutions, it is true that $sqrt[4]{frac{7+3sqrt{5}}{2}}=frac{1+sqrt{5}}{2}.$
        $endgroup$
        – bluemaster
        Dec 22 '18 at 1:08











      Your Answer





      StackExchange.ifUsing("editor", function () {
      return StackExchange.using("mathjaxEditing", function () {
      StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
      StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
      });
      });
      }, "mathjax-editing");

      StackExchange.ready(function() {
      var channelOptions = {
      tags: "".split(" "),
      id: "69"
      };
      initTagRenderer("".split(" "), "".split(" "), channelOptions);

      StackExchange.using("externalEditor", function() {
      // Have to fire editor after snippets, if snippets enabled
      if (StackExchange.settings.snippets.snippetsEnabled) {
      StackExchange.using("snippets", function() {
      createEditor();
      });
      }
      else {
      createEditor();
      }
      });

      function createEditor() {
      StackExchange.prepareEditor({
      heartbeatType: 'answer',
      autoActivateHeartbeat: false,
      convertImagesToLinks: true,
      noModals: true,
      showLowRepImageUploadWarning: true,
      reputationToPostImages: 10,
      bindNavPrevention: true,
      postfix: "",
      imageUploader: {
      brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
      contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
      allowUrls: true
      },
      noCode: true, onDemand: true,
      discardSelector: ".discard-answer"
      ,immediatelyShowMarkdownHelp:true
      });


      }
      });














      draft saved

      draft discarded


















      StackExchange.ready(
      function () {
      StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3048864%2fwith-z-in-mathbb-c-find-the-maximum-value-for-lvert-z-rvert-such-that-lv%23new-answer', 'question_page');
      }
      );

      Post as a guest















      Required, but never shown

























      6 Answers
      6






      active

      oldest

      votes








      6 Answers
      6






      active

      oldest

      votes









      active

      oldest

      votes






      active

      oldest

      votes









      1












      $begingroup$

      Since $$|z| = |z^2-(-1)|geq |z|^2-|-1|$$



      Write $r=|z|$ and we have $$r^2-r-1leq 0implies rin [0,r_2]$$ where $$r_{1,2} ={1pm sqrt{5}over 2} $$



      So $rleq displaystyle{1+ sqrt{5}over 2}$






      share|cite|improve this answer









      $endgroup$


















        1












        $begingroup$

        Since $$|z| = |z^2-(-1)|geq |z|^2-|-1|$$



        Write $r=|z|$ and we have $$r^2-r-1leq 0implies rin [0,r_2]$$ where $$r_{1,2} ={1pm sqrt{5}over 2} $$



        So $rleq displaystyle{1+ sqrt{5}over 2}$






        share|cite|improve this answer









        $endgroup$
















          1












          1








          1





          $begingroup$

          Since $$|z| = |z^2-(-1)|geq |z|^2-|-1|$$



          Write $r=|z|$ and we have $$r^2-r-1leq 0implies rin [0,r_2]$$ where $$r_{1,2} ={1pm sqrt{5}over 2} $$



          So $rleq displaystyle{1+ sqrt{5}over 2}$






          share|cite|improve this answer









          $endgroup$



          Since $$|z| = |z^2-(-1)|geq |z|^2-|-1|$$



          Write $r=|z|$ and we have $$r^2-r-1leq 0implies rin [0,r_2]$$ where $$r_{1,2} ={1pm sqrt{5}over 2} $$



          So $rleq displaystyle{1+ sqrt{5}over 2}$







          share|cite|improve this answer












          share|cite|improve this answer



          share|cite|improve this answer










          answered Dec 21 '18 at 20:57









          greedoidgreedoid

          43.3k1153106




          43.3k1153106























              1












              $begingroup$

              The reverse triangle inequality implies that $|z^2+1|geq ||z^2|-1|$ so that if $|z|^2geq 1$ we see that
              $$|z|geq |z|^2-1$$
              By solving this inequality using the quadratic formula, we see that $|z|leq frac{1}{2}(1+sqrt{5})$, so the maximum must be less than $frac{1}{2}(1+sqrt{5})$. We see that this maximum is attained by setting $z=frac{1}{2}(1+sqrt{5})$.






              share|cite|improve this answer









              $endgroup$













              • $begingroup$
                If $z = frac{1}{2}(1 + sqrt{5})$ then $z^{-1} = frac{1}{2}(-1 + sqrt{5})$ and $|z + z^{-1}| = sqrt{5} ne 1$. The maximum is attained when $z = pm frac{1}{2}(1 + sqrt{5}) i$.
                $endgroup$
                – Daniel Schepler
                Dec 21 '18 at 22:08


















              1












              $begingroup$

              The reverse triangle inequality implies that $|z^2+1|geq ||z^2|-1|$ so that if $|z|^2geq 1$ we see that
              $$|z|geq |z|^2-1$$
              By solving this inequality using the quadratic formula, we see that $|z|leq frac{1}{2}(1+sqrt{5})$, so the maximum must be less than $frac{1}{2}(1+sqrt{5})$. We see that this maximum is attained by setting $z=frac{1}{2}(1+sqrt{5})$.






              share|cite|improve this answer









              $endgroup$













              • $begingroup$
                If $z = frac{1}{2}(1 + sqrt{5})$ then $z^{-1} = frac{1}{2}(-1 + sqrt{5})$ and $|z + z^{-1}| = sqrt{5} ne 1$. The maximum is attained when $z = pm frac{1}{2}(1 + sqrt{5}) i$.
                $endgroup$
                – Daniel Schepler
                Dec 21 '18 at 22:08
















              1












              1








              1





              $begingroup$

              The reverse triangle inequality implies that $|z^2+1|geq ||z^2|-1|$ so that if $|z|^2geq 1$ we see that
              $$|z|geq |z|^2-1$$
              By solving this inequality using the quadratic formula, we see that $|z|leq frac{1}{2}(1+sqrt{5})$, so the maximum must be less than $frac{1}{2}(1+sqrt{5})$. We see that this maximum is attained by setting $z=frac{1}{2}(1+sqrt{5})$.






              share|cite|improve this answer









              $endgroup$



              The reverse triangle inequality implies that $|z^2+1|geq ||z^2|-1|$ so that if $|z|^2geq 1$ we see that
              $$|z|geq |z|^2-1$$
              By solving this inequality using the quadratic formula, we see that $|z|leq frac{1}{2}(1+sqrt{5})$, so the maximum must be less than $frac{1}{2}(1+sqrt{5})$. We see that this maximum is attained by setting $z=frac{1}{2}(1+sqrt{5})$.







              share|cite|improve this answer












              share|cite|improve this answer



              share|cite|improve this answer










              answered Dec 21 '18 at 20:58









              user293794user293794

              1,711613




              1,711613












              • $begingroup$
                If $z = frac{1}{2}(1 + sqrt{5})$ then $z^{-1} = frac{1}{2}(-1 + sqrt{5})$ and $|z + z^{-1}| = sqrt{5} ne 1$. The maximum is attained when $z = pm frac{1}{2}(1 + sqrt{5}) i$.
                $endgroup$
                – Daniel Schepler
                Dec 21 '18 at 22:08




















              • $begingroup$
                If $z = frac{1}{2}(1 + sqrt{5})$ then $z^{-1} = frac{1}{2}(-1 + sqrt{5})$ and $|z + z^{-1}| = sqrt{5} ne 1$. The maximum is attained when $z = pm frac{1}{2}(1 + sqrt{5}) i$.
                $endgroup$
                – Daniel Schepler
                Dec 21 '18 at 22:08


















              $begingroup$
              If $z = frac{1}{2}(1 + sqrt{5})$ then $z^{-1} = frac{1}{2}(-1 + sqrt{5})$ and $|z + z^{-1}| = sqrt{5} ne 1$. The maximum is attained when $z = pm frac{1}{2}(1 + sqrt{5}) i$.
              $endgroup$
              – Daniel Schepler
              Dec 21 '18 at 22:08






              $begingroup$
              If $z = frac{1}{2}(1 + sqrt{5})$ then $z^{-1} = frac{1}{2}(-1 + sqrt{5})$ and $|z + z^{-1}| = sqrt{5} ne 1$. The maximum is attained when $z = pm frac{1}{2}(1 + sqrt{5}) i$.
              $endgroup$
              – Daniel Schepler
              Dec 21 '18 at 22:08













              1












              $begingroup$

              Here's a solution using the dual problem. Assume that the solution of the problem occurs at $z = re^{itheta}.$ Consider
              $$
              F_r(t) = |re^{it}+r^{-1}e^{-it}|^2 = r^2 + r^{-2} + e^{2it}+e^{-2it} = r^2+r^{-2}+2cos 2t.
              $$
              If $z= re^{itheta}$ is the maximizer, we claim that $F_r'(theta)=0$. That is, given $|z|=r$, the angle $theta$ should be chosen so that $F_r(t)$ is either maximized or minimized. To see this, assume to the contrary that $F_r'(theta)neq 0$. Then, there is small $u$ (we allow $u$ to have negative sign) such that $$F_r(theta-u)<1=F(theta)<F_r(theta+u).$$ If it is the case that $rleq 1$, then by choosing slightly large $r'$, we can make $F_{r'}(theta+u)=1$. This contradicts maximality of $r$. If $r>1$, then by choosing slightly large $r'$, we can make $F_{r'}(theta-u) = 1$. This also leads to a contradiction. Now it follows that $cos(2theta)$ should be either $-1$ or $1$. However, we can easily see that $cos(2theta)$ cannot be $1$. Thus, from $cos(2theta)=-1$, we have
              $$
              1=F_r(theta)=r^2 +r^{-2} -2 = (r-r^{-1})^2,
              $$
              or
              $$
              r-1/r = pm 1.
              $$
              This gives
              $$
              r= frac{pm 1+ sqrt{5}}{2}.
              $$
              Therefore, the larger one $r=frac{ 1+ sqrt{5}}{2}$ is the solution.






              share|cite|improve this answer









              $endgroup$


















                1












                $begingroup$

                Here's a solution using the dual problem. Assume that the solution of the problem occurs at $z = re^{itheta}.$ Consider
                $$
                F_r(t) = |re^{it}+r^{-1}e^{-it}|^2 = r^2 + r^{-2} + e^{2it}+e^{-2it} = r^2+r^{-2}+2cos 2t.
                $$
                If $z= re^{itheta}$ is the maximizer, we claim that $F_r'(theta)=0$. That is, given $|z|=r$, the angle $theta$ should be chosen so that $F_r(t)$ is either maximized or minimized. To see this, assume to the contrary that $F_r'(theta)neq 0$. Then, there is small $u$ (we allow $u$ to have negative sign) such that $$F_r(theta-u)<1=F(theta)<F_r(theta+u).$$ If it is the case that $rleq 1$, then by choosing slightly large $r'$, we can make $F_{r'}(theta+u)=1$. This contradicts maximality of $r$. If $r>1$, then by choosing slightly large $r'$, we can make $F_{r'}(theta-u) = 1$. This also leads to a contradiction. Now it follows that $cos(2theta)$ should be either $-1$ or $1$. However, we can easily see that $cos(2theta)$ cannot be $1$. Thus, from $cos(2theta)=-1$, we have
                $$
                1=F_r(theta)=r^2 +r^{-2} -2 = (r-r^{-1})^2,
                $$
                or
                $$
                r-1/r = pm 1.
                $$
                This gives
                $$
                r= frac{pm 1+ sqrt{5}}{2}.
                $$
                Therefore, the larger one $r=frac{ 1+ sqrt{5}}{2}$ is the solution.






                share|cite|improve this answer









                $endgroup$
















                  1












                  1








                  1





                  $begingroup$

                  Here's a solution using the dual problem. Assume that the solution of the problem occurs at $z = re^{itheta}.$ Consider
                  $$
                  F_r(t) = |re^{it}+r^{-1}e^{-it}|^2 = r^2 + r^{-2} + e^{2it}+e^{-2it} = r^2+r^{-2}+2cos 2t.
                  $$
                  If $z= re^{itheta}$ is the maximizer, we claim that $F_r'(theta)=0$. That is, given $|z|=r$, the angle $theta$ should be chosen so that $F_r(t)$ is either maximized or minimized. To see this, assume to the contrary that $F_r'(theta)neq 0$. Then, there is small $u$ (we allow $u$ to have negative sign) such that $$F_r(theta-u)<1=F(theta)<F_r(theta+u).$$ If it is the case that $rleq 1$, then by choosing slightly large $r'$, we can make $F_{r'}(theta+u)=1$. This contradicts maximality of $r$. If $r>1$, then by choosing slightly large $r'$, we can make $F_{r'}(theta-u) = 1$. This also leads to a contradiction. Now it follows that $cos(2theta)$ should be either $-1$ or $1$. However, we can easily see that $cos(2theta)$ cannot be $1$. Thus, from $cos(2theta)=-1$, we have
                  $$
                  1=F_r(theta)=r^2 +r^{-2} -2 = (r-r^{-1})^2,
                  $$
                  or
                  $$
                  r-1/r = pm 1.
                  $$
                  This gives
                  $$
                  r= frac{pm 1+ sqrt{5}}{2}.
                  $$
                  Therefore, the larger one $r=frac{ 1+ sqrt{5}}{2}$ is the solution.






                  share|cite|improve this answer









                  $endgroup$



                  Here's a solution using the dual problem. Assume that the solution of the problem occurs at $z = re^{itheta}.$ Consider
                  $$
                  F_r(t) = |re^{it}+r^{-1}e^{-it}|^2 = r^2 + r^{-2} + e^{2it}+e^{-2it} = r^2+r^{-2}+2cos 2t.
                  $$
                  If $z= re^{itheta}$ is the maximizer, we claim that $F_r'(theta)=0$. That is, given $|z|=r$, the angle $theta$ should be chosen so that $F_r(t)$ is either maximized or minimized. To see this, assume to the contrary that $F_r'(theta)neq 0$. Then, there is small $u$ (we allow $u$ to have negative sign) such that $$F_r(theta-u)<1=F(theta)<F_r(theta+u).$$ If it is the case that $rleq 1$, then by choosing slightly large $r'$, we can make $F_{r'}(theta+u)=1$. This contradicts maximality of $r$. If $r>1$, then by choosing slightly large $r'$, we can make $F_{r'}(theta-u) = 1$. This also leads to a contradiction. Now it follows that $cos(2theta)$ should be either $-1$ or $1$. However, we can easily see that $cos(2theta)$ cannot be $1$. Thus, from $cos(2theta)=-1$, we have
                  $$
                  1=F_r(theta)=r^2 +r^{-2} -2 = (r-r^{-1})^2,
                  $$
                  or
                  $$
                  r-1/r = pm 1.
                  $$
                  This gives
                  $$
                  r= frac{pm 1+ sqrt{5}}{2}.
                  $$
                  Therefore, the larger one $r=frac{ 1+ sqrt{5}}{2}$ is the solution.







                  share|cite|improve this answer












                  share|cite|improve this answer



                  share|cite|improve this answer










                  answered Dec 21 '18 at 21:25









                  SongSong

                  14.2k1633




                  14.2k1633























                      1












                      $begingroup$

                      Suppose $z + frac{1}{z} = e^{it}$ for some $t in mathbb{R}$. Then solving for $z$ gives
                      $$z = frac{e^{it} pm sqrt{e^{2it} - 4}}{2} = e^{it} cdot frac{1 pm sqrt{1 - 4 e^{-2it}}}{2}.$$
                      Therefore,
                      $$ |z| = frac{1}{2} left| 1 pm sqrt{1 - 4 e^{-2it}} right| le frac{1}{2} left( |1| + |sqrt{1 - 4 e^{-2it}}|right) = frac{1}{2} left( 1 + sqrt{1 - 4 e^{-2it}} right)le \
                      frac{1}{2} left( 1 + sqrt{|1| + |4 e^{-2it}|}right) = frac{1 + sqrt{5}}{2}.$$

                      On the other hand, if we set $t := frac{pi}{2}$ so that $e^{-2it} = -1$ then we will get equality. Therefore, the maximum value of $|z|$ is $frac{1+sqrt{5}}{2}$, which is achieved for $z = frac{1+sqrt{5}}{2} i$. (In fact, a slight refinement of the argument will show that equality holds, i.e. the maximum is achieved, if and only if $z = pm frac{1+sqrt{5}}{2} i$.)






                      share|cite|improve this answer











                      $endgroup$













                      • $begingroup$
                        It seems an interesting approach, can you expand a little bit more... how this can lead to the maximum $|z|$.
                        $endgroup$
                        – bluemaster
                        Dec 21 '18 at 21:26






                      • 1




                        $begingroup$
                        OK, I've filled in how the rest of the argument goes.
                        $endgroup$
                        – Daniel Schepler
                        Dec 21 '18 at 22:07
















                      1












                      $begingroup$

                      Suppose $z + frac{1}{z} = e^{it}$ for some $t in mathbb{R}$. Then solving for $z$ gives
                      $$z = frac{e^{it} pm sqrt{e^{2it} - 4}}{2} = e^{it} cdot frac{1 pm sqrt{1 - 4 e^{-2it}}}{2}.$$
                      Therefore,
                      $$ |z| = frac{1}{2} left| 1 pm sqrt{1 - 4 e^{-2it}} right| le frac{1}{2} left( |1| + |sqrt{1 - 4 e^{-2it}}|right) = frac{1}{2} left( 1 + sqrt{1 - 4 e^{-2it}} right)le \
                      frac{1}{2} left( 1 + sqrt{|1| + |4 e^{-2it}|}right) = frac{1 + sqrt{5}}{2}.$$

                      On the other hand, if we set $t := frac{pi}{2}$ so that $e^{-2it} = -1$ then we will get equality. Therefore, the maximum value of $|z|$ is $frac{1+sqrt{5}}{2}$, which is achieved for $z = frac{1+sqrt{5}}{2} i$. (In fact, a slight refinement of the argument will show that equality holds, i.e. the maximum is achieved, if and only if $z = pm frac{1+sqrt{5}}{2} i$.)






                      share|cite|improve this answer











                      $endgroup$













                      • $begingroup$
                        It seems an interesting approach, can you expand a little bit more... how this can lead to the maximum $|z|$.
                        $endgroup$
                        – bluemaster
                        Dec 21 '18 at 21:26






                      • 1




                        $begingroup$
                        OK, I've filled in how the rest of the argument goes.
                        $endgroup$
                        – Daniel Schepler
                        Dec 21 '18 at 22:07














                      1












                      1








                      1





                      $begingroup$

                      Suppose $z + frac{1}{z} = e^{it}$ for some $t in mathbb{R}$. Then solving for $z$ gives
                      $$z = frac{e^{it} pm sqrt{e^{2it} - 4}}{2} = e^{it} cdot frac{1 pm sqrt{1 - 4 e^{-2it}}}{2}.$$
                      Therefore,
                      $$ |z| = frac{1}{2} left| 1 pm sqrt{1 - 4 e^{-2it}} right| le frac{1}{2} left( |1| + |sqrt{1 - 4 e^{-2it}}|right) = frac{1}{2} left( 1 + sqrt{1 - 4 e^{-2it}} right)le \
                      frac{1}{2} left( 1 + sqrt{|1| + |4 e^{-2it}|}right) = frac{1 + sqrt{5}}{2}.$$

                      On the other hand, if we set $t := frac{pi}{2}$ so that $e^{-2it} = -1$ then we will get equality. Therefore, the maximum value of $|z|$ is $frac{1+sqrt{5}}{2}$, which is achieved for $z = frac{1+sqrt{5}}{2} i$. (In fact, a slight refinement of the argument will show that equality holds, i.e. the maximum is achieved, if and only if $z = pm frac{1+sqrt{5}}{2} i$.)






                      share|cite|improve this answer











                      $endgroup$



                      Suppose $z + frac{1}{z} = e^{it}$ for some $t in mathbb{R}$. Then solving for $z$ gives
                      $$z = frac{e^{it} pm sqrt{e^{2it} - 4}}{2} = e^{it} cdot frac{1 pm sqrt{1 - 4 e^{-2it}}}{2}.$$
                      Therefore,
                      $$ |z| = frac{1}{2} left| 1 pm sqrt{1 - 4 e^{-2it}} right| le frac{1}{2} left( |1| + |sqrt{1 - 4 e^{-2it}}|right) = frac{1}{2} left( 1 + sqrt{1 - 4 e^{-2it}} right)le \
                      frac{1}{2} left( 1 + sqrt{|1| + |4 e^{-2it}|}right) = frac{1 + sqrt{5}}{2}.$$

                      On the other hand, if we set $t := frac{pi}{2}$ so that $e^{-2it} = -1$ then we will get equality. Therefore, the maximum value of $|z|$ is $frac{1+sqrt{5}}{2}$, which is achieved for $z = frac{1+sqrt{5}}{2} i$. (In fact, a slight refinement of the argument will show that equality holds, i.e. the maximum is achieved, if and only if $z = pm frac{1+sqrt{5}}{2} i$.)







                      share|cite|improve this answer














                      share|cite|improve this answer



                      share|cite|improve this answer








                      edited Dec 21 '18 at 22:06

























                      answered Dec 21 '18 at 20:49









                      Daniel ScheplerDaniel Schepler

                      8,8791620




                      8,8791620












                      • $begingroup$
                        It seems an interesting approach, can you expand a little bit more... how this can lead to the maximum $|z|$.
                        $endgroup$
                        – bluemaster
                        Dec 21 '18 at 21:26






                      • 1




                        $begingroup$
                        OK, I've filled in how the rest of the argument goes.
                        $endgroup$
                        – Daniel Schepler
                        Dec 21 '18 at 22:07


















                      • $begingroup$
                        It seems an interesting approach, can you expand a little bit more... how this can lead to the maximum $|z|$.
                        $endgroup$
                        – bluemaster
                        Dec 21 '18 at 21:26






                      • 1




                        $begingroup$
                        OK, I've filled in how the rest of the argument goes.
                        $endgroup$
                        – Daniel Schepler
                        Dec 21 '18 at 22:07
















                      $begingroup$
                      It seems an interesting approach, can you expand a little bit more... how this can lead to the maximum $|z|$.
                      $endgroup$
                      – bluemaster
                      Dec 21 '18 at 21:26




                      $begingroup$
                      It seems an interesting approach, can you expand a little bit more... how this can lead to the maximum $|z|$.
                      $endgroup$
                      – bluemaster
                      Dec 21 '18 at 21:26




                      1




                      1




                      $begingroup$
                      OK, I've filled in how the rest of the argument goes.
                      $endgroup$
                      – Daniel Schepler
                      Dec 21 '18 at 22:07




                      $begingroup$
                      OK, I've filled in how the rest of the argument goes.
                      $endgroup$
                      – Daniel Schepler
                      Dec 21 '18 at 22:07











                      1












                      $begingroup$

                      $|z| =|z+1/z-1/z| le$



                      $ |z+1/z| +|1/z| =1+|1/z|;$



                      $r:=|z|$;



                      $r-1/r le 1;$



                      $r^2-r -1le 0;$



                      $(r-1/2)^2 -1/4-1le 0;$



                      $r-1/2 le (1/2)√5;$



                      $r le (1/2)(1+√5)$.



                      Is the maximum attained?






                      share|cite|improve this answer











                      $endgroup$













                      • $begingroup$
                        Very clear development. What is a simple argument to show that this is not just an upper bound but the maximum?
                        $endgroup$
                        – bluemaster
                        Dec 21 '18 at 22:59










                      • $begingroup$
                        @RoryDaulton: I'm thinking on a math-contest situation, how could we find an appropriate value that leads to this maximum value. It is not obvious. The solution using the optimization approach appears to me that address this difficulty.
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 1:09






                      • 1




                        $begingroup$
                        bluemaster.Easy argument !? I try a bit so easy argument.Triangle inequality in the complex plane: Vector a=z , b=1/z .For side lengths of triangle formed by a,b and a+b (vectors): Since a triangle triangle inequality is strict((>).(lengths of sum of 2 sides of triangle is >than 3rd side.) Equality can only happen if a and b are collinear, We can try z= 1/2(1+√5)(real), does not work, but z= (1/2)(1+√5)i (purely imaginary), works to give |z+1/z|=1.Your thoughts?
                        $endgroup$
                        – Peter Szilas
                        Dec 22 '18 at 7:32










                      • $begingroup$
                        Very good argument. I was struggling to see by the geometric interpretation and your thoughts clarified. The key is the situation in which a and b are collinear. Thanks!
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 9:03










                      • $begingroup$
                        bluemaster.Thanks! I was struggling too:))Greetings.
                        $endgroup$
                        – Peter Szilas
                        Dec 22 '18 at 9:14
















                      1












                      $begingroup$

                      $|z| =|z+1/z-1/z| le$



                      $ |z+1/z| +|1/z| =1+|1/z|;$



                      $r:=|z|$;



                      $r-1/r le 1;$



                      $r^2-r -1le 0;$



                      $(r-1/2)^2 -1/4-1le 0;$



                      $r-1/2 le (1/2)√5;$



                      $r le (1/2)(1+√5)$.



                      Is the maximum attained?






                      share|cite|improve this answer











                      $endgroup$













                      • $begingroup$
                        Very clear development. What is a simple argument to show that this is not just an upper bound but the maximum?
                        $endgroup$
                        – bluemaster
                        Dec 21 '18 at 22:59










                      • $begingroup$
                        @RoryDaulton: I'm thinking on a math-contest situation, how could we find an appropriate value that leads to this maximum value. It is not obvious. The solution using the optimization approach appears to me that address this difficulty.
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 1:09






                      • 1




                        $begingroup$
                        bluemaster.Easy argument !? I try a bit so easy argument.Triangle inequality in the complex plane: Vector a=z , b=1/z .For side lengths of triangle formed by a,b and a+b (vectors): Since a triangle triangle inequality is strict((>).(lengths of sum of 2 sides of triangle is >than 3rd side.) Equality can only happen if a and b are collinear, We can try z= 1/2(1+√5)(real), does not work, but z= (1/2)(1+√5)i (purely imaginary), works to give |z+1/z|=1.Your thoughts?
                        $endgroup$
                        – Peter Szilas
                        Dec 22 '18 at 7:32










                      • $begingroup$
                        Very good argument. I was struggling to see by the geometric interpretation and your thoughts clarified. The key is the situation in which a and b are collinear. Thanks!
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 9:03










                      • $begingroup$
                        bluemaster.Thanks! I was struggling too:))Greetings.
                        $endgroup$
                        – Peter Szilas
                        Dec 22 '18 at 9:14














                      1












                      1








                      1





                      $begingroup$

                      $|z| =|z+1/z-1/z| le$



                      $ |z+1/z| +|1/z| =1+|1/z|;$



                      $r:=|z|$;



                      $r-1/r le 1;$



                      $r^2-r -1le 0;$



                      $(r-1/2)^2 -1/4-1le 0;$



                      $r-1/2 le (1/2)√5;$



                      $r le (1/2)(1+√5)$.



                      Is the maximum attained?






                      share|cite|improve this answer











                      $endgroup$



                      $|z| =|z+1/z-1/z| le$



                      $ |z+1/z| +|1/z| =1+|1/z|;$



                      $r:=|z|$;



                      $r-1/r le 1;$



                      $r^2-r -1le 0;$



                      $(r-1/2)^2 -1/4-1le 0;$



                      $r-1/2 le (1/2)√5;$



                      $r le (1/2)(1+√5)$.



                      Is the maximum attained?







                      share|cite|improve this answer














                      share|cite|improve this answer



                      share|cite|improve this answer








                      edited Dec 22 '18 at 7:06

























                      answered Dec 21 '18 at 22:35









                      Peter SzilasPeter Szilas

                      11.3k2822




                      11.3k2822












                      • $begingroup$
                        Very clear development. What is a simple argument to show that this is not just an upper bound but the maximum?
                        $endgroup$
                        – bluemaster
                        Dec 21 '18 at 22:59










                      • $begingroup$
                        @RoryDaulton: I'm thinking on a math-contest situation, how could we find an appropriate value that leads to this maximum value. It is not obvious. The solution using the optimization approach appears to me that address this difficulty.
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 1:09






                      • 1




                        $begingroup$
                        bluemaster.Easy argument !? I try a bit so easy argument.Triangle inequality in the complex plane: Vector a=z , b=1/z .For side lengths of triangle formed by a,b and a+b (vectors): Since a triangle triangle inequality is strict((>).(lengths of sum of 2 sides of triangle is >than 3rd side.) Equality can only happen if a and b are collinear, We can try z= 1/2(1+√5)(real), does not work, but z= (1/2)(1+√5)i (purely imaginary), works to give |z+1/z|=1.Your thoughts?
                        $endgroup$
                        – Peter Szilas
                        Dec 22 '18 at 7:32










                      • $begingroup$
                        Very good argument. I was struggling to see by the geometric interpretation and your thoughts clarified. The key is the situation in which a and b are collinear. Thanks!
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 9:03










                      • $begingroup$
                        bluemaster.Thanks! I was struggling too:))Greetings.
                        $endgroup$
                        – Peter Szilas
                        Dec 22 '18 at 9:14


















                      • $begingroup$
                        Very clear development. What is a simple argument to show that this is not just an upper bound but the maximum?
                        $endgroup$
                        – bluemaster
                        Dec 21 '18 at 22:59










                      • $begingroup$
                        @RoryDaulton: I'm thinking on a math-contest situation, how could we find an appropriate value that leads to this maximum value. It is not obvious. The solution using the optimization approach appears to me that address this difficulty.
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 1:09






                      • 1




                        $begingroup$
                        bluemaster.Easy argument !? I try a bit so easy argument.Triangle inequality in the complex plane: Vector a=z , b=1/z .For side lengths of triangle formed by a,b and a+b (vectors): Since a triangle triangle inequality is strict((>).(lengths of sum of 2 sides of triangle is >than 3rd side.) Equality can only happen if a and b are collinear, We can try z= 1/2(1+√5)(real), does not work, but z= (1/2)(1+√5)i (purely imaginary), works to give |z+1/z|=1.Your thoughts?
                        $endgroup$
                        – Peter Szilas
                        Dec 22 '18 at 7:32










                      • $begingroup$
                        Very good argument. I was struggling to see by the geometric interpretation and your thoughts clarified. The key is the situation in which a and b are collinear. Thanks!
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 9:03










                      • $begingroup$
                        bluemaster.Thanks! I was struggling too:))Greetings.
                        $endgroup$
                        – Peter Szilas
                        Dec 22 '18 at 9:14
















                      $begingroup$
                      Very clear development. What is a simple argument to show that this is not just an upper bound but the maximum?
                      $endgroup$
                      – bluemaster
                      Dec 21 '18 at 22:59




                      $begingroup$
                      Very clear development. What is a simple argument to show that this is not just an upper bound but the maximum?
                      $endgroup$
                      – bluemaster
                      Dec 21 '18 at 22:59












                      $begingroup$
                      @RoryDaulton: I'm thinking on a math-contest situation, how could we find an appropriate value that leads to this maximum value. It is not obvious. The solution using the optimization approach appears to me that address this difficulty.
                      $endgroup$
                      – bluemaster
                      Dec 22 '18 at 1:09




                      $begingroup$
                      @RoryDaulton: I'm thinking on a math-contest situation, how could we find an appropriate value that leads to this maximum value. It is not obvious. The solution using the optimization approach appears to me that address this difficulty.
                      $endgroup$
                      – bluemaster
                      Dec 22 '18 at 1:09




                      1




                      1




                      $begingroup$
                      bluemaster.Easy argument !? I try a bit so easy argument.Triangle inequality in the complex plane: Vector a=z , b=1/z .For side lengths of triangle formed by a,b and a+b (vectors): Since a triangle triangle inequality is strict((>).(lengths of sum of 2 sides of triangle is >than 3rd side.) Equality can only happen if a and b are collinear, We can try z= 1/2(1+√5)(real), does not work, but z= (1/2)(1+√5)i (purely imaginary), works to give |z+1/z|=1.Your thoughts?
                      $endgroup$
                      – Peter Szilas
                      Dec 22 '18 at 7:32




                      $begingroup$
                      bluemaster.Easy argument !? I try a bit so easy argument.Triangle inequality in the complex plane: Vector a=z , b=1/z .For side lengths of triangle formed by a,b and a+b (vectors): Since a triangle triangle inequality is strict((>).(lengths of sum of 2 sides of triangle is >than 3rd side.) Equality can only happen if a and b are collinear, We can try z= 1/2(1+√5)(real), does not work, but z= (1/2)(1+√5)i (purely imaginary), works to give |z+1/z|=1.Your thoughts?
                      $endgroup$
                      – Peter Szilas
                      Dec 22 '18 at 7:32












                      $begingroup$
                      Very good argument. I was struggling to see by the geometric interpretation and your thoughts clarified. The key is the situation in which a and b are collinear. Thanks!
                      $endgroup$
                      – bluemaster
                      Dec 22 '18 at 9:03




                      $begingroup$
                      Very good argument. I was struggling to see by the geometric interpretation and your thoughts clarified. The key is the situation in which a and b are collinear. Thanks!
                      $endgroup$
                      – bluemaster
                      Dec 22 '18 at 9:03












                      $begingroup$
                      bluemaster.Thanks! I was struggling too:))Greetings.
                      $endgroup$
                      – Peter Szilas
                      Dec 22 '18 at 9:14




                      $begingroup$
                      bluemaster.Thanks! I was struggling too:))Greetings.
                      $endgroup$
                      – Peter Szilas
                      Dec 22 '18 at 9:14











                      0












                      $begingroup$

                      We can maximize $g(a,b) = a^2+b^2$ subject to $(a^2+b^2)^2 =3b^2-1$. Solve for $a^2$ we have: $a^2= sqrt{3b^2-1} -b^2implies g(a,b) = sqrt{3b^2-1}$. Thus the problem now is the find the maximum value of $b^2$. To this end, let $c = b^2 ge 0$, we have: $a^4+2a^2c+c^2 - 3c+ 1 = 0implies c^2-3c+1 le 0implies c le dfrac{3+sqrt{5}}{2}implies 3b^2 -1 = 3c-1le dfrac{7+3sqrt{5}}{2}implies |z|_{text{max}} = sqrt{g(a,b)}_{text{max}} = sqrt[4]{dfrac{7+3sqrt{5}}{2}}= dfrac{1+sqrt{5}}{2}$,and this max occurs when $a = 0$.






                      share|cite|improve this answer











                      $endgroup$













                      • $begingroup$
                        And just to compare with other solutions, it is true that $sqrt[4]{frac{7+3sqrt{5}}{2}}=frac{1+sqrt{5}}{2}.$
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 1:08
















                      0












                      $begingroup$

                      We can maximize $g(a,b) = a^2+b^2$ subject to $(a^2+b^2)^2 =3b^2-1$. Solve for $a^2$ we have: $a^2= sqrt{3b^2-1} -b^2implies g(a,b) = sqrt{3b^2-1}$. Thus the problem now is the find the maximum value of $b^2$. To this end, let $c = b^2 ge 0$, we have: $a^4+2a^2c+c^2 - 3c+ 1 = 0implies c^2-3c+1 le 0implies c le dfrac{3+sqrt{5}}{2}implies 3b^2 -1 = 3c-1le dfrac{7+3sqrt{5}}{2}implies |z|_{text{max}} = sqrt{g(a,b)}_{text{max}} = sqrt[4]{dfrac{7+3sqrt{5}}{2}}= dfrac{1+sqrt{5}}{2}$,and this max occurs when $a = 0$.






                      share|cite|improve this answer











                      $endgroup$













                      • $begingroup$
                        And just to compare with other solutions, it is true that $sqrt[4]{frac{7+3sqrt{5}}{2}}=frac{1+sqrt{5}}{2}.$
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 1:08














                      0












                      0








                      0





                      $begingroup$

                      We can maximize $g(a,b) = a^2+b^2$ subject to $(a^2+b^2)^2 =3b^2-1$. Solve for $a^2$ we have: $a^2= sqrt{3b^2-1} -b^2implies g(a,b) = sqrt{3b^2-1}$. Thus the problem now is the find the maximum value of $b^2$. To this end, let $c = b^2 ge 0$, we have: $a^4+2a^2c+c^2 - 3c+ 1 = 0implies c^2-3c+1 le 0implies c le dfrac{3+sqrt{5}}{2}implies 3b^2 -1 = 3c-1le dfrac{7+3sqrt{5}}{2}implies |z|_{text{max}} = sqrt{g(a,b)}_{text{max}} = sqrt[4]{dfrac{7+3sqrt{5}}{2}}= dfrac{1+sqrt{5}}{2}$,and this max occurs when $a = 0$.






                      share|cite|improve this answer











                      $endgroup$



                      We can maximize $g(a,b) = a^2+b^2$ subject to $(a^2+b^2)^2 =3b^2-1$. Solve for $a^2$ we have: $a^2= sqrt{3b^2-1} -b^2implies g(a,b) = sqrt{3b^2-1}$. Thus the problem now is the find the maximum value of $b^2$. To this end, let $c = b^2 ge 0$, we have: $a^4+2a^2c+c^2 - 3c+ 1 = 0implies c^2-3c+1 le 0implies c le dfrac{3+sqrt{5}}{2}implies 3b^2 -1 = 3c-1le dfrac{7+3sqrt{5}}{2}implies |z|_{text{max}} = sqrt{g(a,b)}_{text{max}} = sqrt[4]{dfrac{7+3sqrt{5}}{2}}= dfrac{1+sqrt{5}}{2}$,and this max occurs when $a = 0$.







                      share|cite|improve this answer














                      share|cite|improve this answer



                      share|cite|improve this answer








                      edited Dec 22 '18 at 8:57

























                      answered Dec 21 '18 at 21:11









                      DeepSeaDeepSea

                      71.3k54487




                      71.3k54487












                      • $begingroup$
                        And just to compare with other solutions, it is true that $sqrt[4]{frac{7+3sqrt{5}}{2}}=frac{1+sqrt{5}}{2}.$
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 1:08


















                      • $begingroup$
                        And just to compare with other solutions, it is true that $sqrt[4]{frac{7+3sqrt{5}}{2}}=frac{1+sqrt{5}}{2}.$
                        $endgroup$
                        – bluemaster
                        Dec 22 '18 at 1:08
















                      $begingroup$
                      And just to compare with other solutions, it is true that $sqrt[4]{frac{7+3sqrt{5}}{2}}=frac{1+sqrt{5}}{2}.$
                      $endgroup$
                      – bluemaster
                      Dec 22 '18 at 1:08




                      $begingroup$
                      And just to compare with other solutions, it is true that $sqrt[4]{frac{7+3sqrt{5}}{2}}=frac{1+sqrt{5}}{2}.$
                      $endgroup$
                      – bluemaster
                      Dec 22 '18 at 1:08


















                      draft saved

                      draft discarded




















































                      Thanks for contributing an answer to Mathematics Stack Exchange!


                      • Please be sure to answer the question. Provide details and share your research!

                      But avoid



                      • Asking for help, clarification, or responding to other answers.

                      • Making statements based on opinion; back them up with references or personal experience.


                      Use MathJax to format equations. MathJax reference.


                      To learn more, see our tips on writing great answers.




                      draft saved


                      draft discarded














                      StackExchange.ready(
                      function () {
                      StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3048864%2fwith-z-in-mathbb-c-find-the-maximum-value-for-lvert-z-rvert-such-that-lv%23new-answer', 'question_page');
                      }
                      );

                      Post as a guest















                      Required, but never shown





















































                      Required, but never shown














                      Required, but never shown












                      Required, but never shown







                      Required, but never shown

































                      Required, but never shown














                      Required, but never shown












                      Required, but never shown







                      Required, but never shown







                      Popular posts from this blog

                      Quarter-circle Tiles

                      build a pushdown automaton that recognizes the reverse language of a given pushdown automaton?

                      Mont Emei