Inequality for three variables












2












$begingroup$


Hello I would like to solve this with $x,y,z$ positive real numbers :



$$sqrt{dfrac{5}{13}}frac{1}{13}dfrac{(x)^3}{(x^2+1)}+dfrac{1}{13}dfrac{(y)^2}{(y^2+1)}+sqrt{dfrac{13}{5}}dfrac{1}{13(y)}dfrac{(z)^2}{(z^2+1)}$$$$geq dfrac{1+sqrt{dfrac{5}{13}}x+sqrt{dfrac{13}{5}}dfrac{1}{y}}{18}$$



under the condition $xyz=(sqrt{frac{13}{5}})^3$



I have no idea to prove this.
Thanks a lot.










share|cite|improve this question











$endgroup$












  • $begingroup$
    Does that first term have $x^3$ or $x^2$? The other two terms have $ ^2$.
    $endgroup$
    – marty cohen
    Oct 20 '17 at 18:05










  • $begingroup$
    It's $x^3$ to have an equality when $x=y=z$.And yes in fact that seems to be strange .Furthermore if we study the following function :$$F(x)=sqrt{dfrac{5}{13}}frac{1}{13}dfrac{(x)^3}{(x^2+1)}+dfrac{1}{13}dfrac{(y)^2}{(y^2+1)}+sqrt{dfrac{13}{5}}dfrac{1}{13(y)}dfrac{(z)^2}{(z^2+1)}-dfrac{1+sqrt{dfrac{5}{13}}x+sqrt{dfrac{13}{5}}dfrac{1}{y}}{18}$$ .We have just to find a minimum and use it with the condition to find an inequality with one variable .But perhaps there exists a simpler method. Thanks for your interest
    $endgroup$
    – user448747
    Oct 20 '17 at 18:24












  • $begingroup$
    @martycohen Furthermore if you prove that the inequality of this link follows :).
    $endgroup$
    – user448747
    Oct 20 '17 at 18:26
















2












$begingroup$


Hello I would like to solve this with $x,y,z$ positive real numbers :



$$sqrt{dfrac{5}{13}}frac{1}{13}dfrac{(x)^3}{(x^2+1)}+dfrac{1}{13}dfrac{(y)^2}{(y^2+1)}+sqrt{dfrac{13}{5}}dfrac{1}{13(y)}dfrac{(z)^2}{(z^2+1)}$$$$geq dfrac{1+sqrt{dfrac{5}{13}}x+sqrt{dfrac{13}{5}}dfrac{1}{y}}{18}$$



under the condition $xyz=(sqrt{frac{13}{5}})^3$



I have no idea to prove this.
Thanks a lot.










share|cite|improve this question











$endgroup$












  • $begingroup$
    Does that first term have $x^3$ or $x^2$? The other two terms have $ ^2$.
    $endgroup$
    – marty cohen
    Oct 20 '17 at 18:05










  • $begingroup$
    It's $x^3$ to have an equality when $x=y=z$.And yes in fact that seems to be strange .Furthermore if we study the following function :$$F(x)=sqrt{dfrac{5}{13}}frac{1}{13}dfrac{(x)^3}{(x^2+1)}+dfrac{1}{13}dfrac{(y)^2}{(y^2+1)}+sqrt{dfrac{13}{5}}dfrac{1}{13(y)}dfrac{(z)^2}{(z^2+1)}-dfrac{1+sqrt{dfrac{5}{13}}x+sqrt{dfrac{13}{5}}dfrac{1}{y}}{18}$$ .We have just to find a minimum and use it with the condition to find an inequality with one variable .But perhaps there exists a simpler method. Thanks for your interest
    $endgroup$
    – user448747
    Oct 20 '17 at 18:24












  • $begingroup$
    @martycohen Furthermore if you prove that the inequality of this link follows :).
    $endgroup$
    – user448747
    Oct 20 '17 at 18:26














2












2








2


1



$begingroup$


Hello I would like to solve this with $x,y,z$ positive real numbers :



$$sqrt{dfrac{5}{13}}frac{1}{13}dfrac{(x)^3}{(x^2+1)}+dfrac{1}{13}dfrac{(y)^2}{(y^2+1)}+sqrt{dfrac{13}{5}}dfrac{1}{13(y)}dfrac{(z)^2}{(z^2+1)}$$$$geq dfrac{1+sqrt{dfrac{5}{13}}x+sqrt{dfrac{13}{5}}dfrac{1}{y}}{18}$$



under the condition $xyz=(sqrt{frac{13}{5}})^3$



I have no idea to prove this.
Thanks a lot.










share|cite|improve this question











$endgroup$




Hello I would like to solve this with $x,y,z$ positive real numbers :



$$sqrt{dfrac{5}{13}}frac{1}{13}dfrac{(x)^3}{(x^2+1)}+dfrac{1}{13}dfrac{(y)^2}{(y^2+1)}+sqrt{dfrac{13}{5}}dfrac{1}{13(y)}dfrac{(z)^2}{(z^2+1)}$$$$geq dfrac{1+sqrt{dfrac{5}{13}}x+sqrt{dfrac{13}{5}}dfrac{1}{y}}{18}$$



under the condition $xyz=(sqrt{frac{13}{5}})^3$



I have no idea to prove this.
Thanks a lot.







real-analysis inequality






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Dec 8 '18 at 11:39









amWhy

1




1










asked Oct 20 '17 at 11:39







user448747



















  • $begingroup$
    Does that first term have $x^3$ or $x^2$? The other two terms have $ ^2$.
    $endgroup$
    – marty cohen
    Oct 20 '17 at 18:05










  • $begingroup$
    It's $x^3$ to have an equality when $x=y=z$.And yes in fact that seems to be strange .Furthermore if we study the following function :$$F(x)=sqrt{dfrac{5}{13}}frac{1}{13}dfrac{(x)^3}{(x^2+1)}+dfrac{1}{13}dfrac{(y)^2}{(y^2+1)}+sqrt{dfrac{13}{5}}dfrac{1}{13(y)}dfrac{(z)^2}{(z^2+1)}-dfrac{1+sqrt{dfrac{5}{13}}x+sqrt{dfrac{13}{5}}dfrac{1}{y}}{18}$$ .We have just to find a minimum and use it with the condition to find an inequality with one variable .But perhaps there exists a simpler method. Thanks for your interest
    $endgroup$
    – user448747
    Oct 20 '17 at 18:24












  • $begingroup$
    @martycohen Furthermore if you prove that the inequality of this link follows :).
    $endgroup$
    – user448747
    Oct 20 '17 at 18:26


















  • $begingroup$
    Does that first term have $x^3$ or $x^2$? The other two terms have $ ^2$.
    $endgroup$
    – marty cohen
    Oct 20 '17 at 18:05










  • $begingroup$
    It's $x^3$ to have an equality when $x=y=z$.And yes in fact that seems to be strange .Furthermore if we study the following function :$$F(x)=sqrt{dfrac{5}{13}}frac{1}{13}dfrac{(x)^3}{(x^2+1)}+dfrac{1}{13}dfrac{(y)^2}{(y^2+1)}+sqrt{dfrac{13}{5}}dfrac{1}{13(y)}dfrac{(z)^2}{(z^2+1)}-dfrac{1+sqrt{dfrac{5}{13}}x+sqrt{dfrac{13}{5}}dfrac{1}{y}}{18}$$ .We have just to find a minimum and use it with the condition to find an inequality with one variable .But perhaps there exists a simpler method. Thanks for your interest
    $endgroup$
    – user448747
    Oct 20 '17 at 18:24












  • $begingroup$
    @martycohen Furthermore if you prove that the inequality of this link follows :).
    $endgroup$
    – user448747
    Oct 20 '17 at 18:26
















$begingroup$
Does that first term have $x^3$ or $x^2$? The other two terms have $ ^2$.
$endgroup$
– marty cohen
Oct 20 '17 at 18:05




$begingroup$
Does that first term have $x^3$ or $x^2$? The other two terms have $ ^2$.
$endgroup$
– marty cohen
Oct 20 '17 at 18:05












$begingroup$
It's $x^3$ to have an equality when $x=y=z$.And yes in fact that seems to be strange .Furthermore if we study the following function :$$F(x)=sqrt{dfrac{5}{13}}frac{1}{13}dfrac{(x)^3}{(x^2+1)}+dfrac{1}{13}dfrac{(y)^2}{(y^2+1)}+sqrt{dfrac{13}{5}}dfrac{1}{13(y)}dfrac{(z)^2}{(z^2+1)}-dfrac{1+sqrt{dfrac{5}{13}}x+sqrt{dfrac{13}{5}}dfrac{1}{y}}{18}$$ .We have just to find a minimum and use it with the condition to find an inequality with one variable .But perhaps there exists a simpler method. Thanks for your interest
$endgroup$
– user448747
Oct 20 '17 at 18:24






$begingroup$
It's $x^3$ to have an equality when $x=y=z$.And yes in fact that seems to be strange .Furthermore if we study the following function :$$F(x)=sqrt{dfrac{5}{13}}frac{1}{13}dfrac{(x)^3}{(x^2+1)}+dfrac{1}{13}dfrac{(y)^2}{(y^2+1)}+sqrt{dfrac{13}{5}}dfrac{1}{13(y)}dfrac{(z)^2}{(z^2+1)}-dfrac{1+sqrt{dfrac{5}{13}}x+sqrt{dfrac{13}{5}}dfrac{1}{y}}{18}$$ .We have just to find a minimum and use it with the condition to find an inequality with one variable .But perhaps there exists a simpler method. Thanks for your interest
$endgroup$
– user448747
Oct 20 '17 at 18:24














$begingroup$
@martycohen Furthermore if you prove that the inequality of this link follows :).
$endgroup$
– user448747
Oct 20 '17 at 18:26




$begingroup$
@martycohen Furthermore if you prove that the inequality of this link follows :).
$endgroup$
– user448747
Oct 20 '17 at 18:26










2 Answers
2






active

oldest

votes


















1












$begingroup$

This answer is extremely technical and uses Mathcad calculations.



By routine calculations the inequality can be simplified to



$$frac{5}{18}left(frac xA +1+frac Ayright)ge frac xAcdot frac 1{x^2+1}+frac 1{y^2+1}+frac Aycdot frac 1{z^2+1},$$



where $A=sqrtfrac{13}5$.



Or, because $z^2=frac {A^6}{x^2y^2}$,



$$frac{5}{18}left(frac xA +1+frac Ayright)ge frac xAcdot frac 1{x^2+1}+frac 1{y^2+1}+frac Aycdot frac { x^2y^2}{A^6 + x^2y^2},$$



Substituting $u=frac xA$, $v=frac yA $ we obtain



$$frac{5}{18}left(u+1+frac 1vright)ge ucdot frac 1{A^2u^2+1}+frac 1{A^2v^2+1}+frac 1vcdot frac {u^2v^2}{A^2 + u^2v^2}$$



$$u+1+frac 1vge 18left(ucdot frac 1{13u^2+5}+frac 1{13v^2+5}+frac 1vcdot frac {u^2v^2}{13 + 5u^2v^2}right)$$



Substituting $w=frac 1v$, we obtain



$$u+1+wge 18left(frac u{13u^2+5}+frac {w^2}{13+5w^2}+ frac {wu^2}{13w^2 + 5u^2}right)$$



$25(u^5w^2+u^2+w^5)+65(u^5-u^4w^3-u^4w^2+u^4+u^3w^4-u^3+u^2w^5-u^2w-uw^4-w^4+w^3+w^2)+144 (u^3w^2+u^2w^3+u^2w^2)-169(u^4w+ u^2w^4+uw^2)ge 0$



The graphs suggest that this inequality is true. We may look for its proof as follows.



Denote the left hand side of the last inequality by $f(u,w)$. We are going to show that $inf{f(u,w): u,wge 0}=0$. When one of the variables $u$ and $w$ is fixed, $f(u,w)$ becomes a polynomial with respect to the other (which we denote by $v$) with the leading coefficient at least $25$. So it attains its minimum when $v=0$ or $frac{partial f}{partial v}=0$. If $u=0$ then $f(u,w)=25w^5-65w^4+65w^3+65w^2=w^3(5w-8)^2+15w^4+w^3+65w^2ge 0$. If $w=0$ then $f(u,w)=65u^5+65u^4-65u^3+25u^2=u^2(8u-5)^2+65u^5+u^4+15u^3ge 0$.



Conditions $frac{partial f}{partial u}=0$ and $frac{partial f}{partial w}=0$ yield the system (*)



$130uw^5+(195u^2-338u-65)w^4+(-260u^3+288u)w^3+(125u^4-260u^3+432u^2+288u-169)w^2+(-676u^3-130u)w+
325u^4+260u^3-195u^2+50u=0$



$(325u^2+125)w^4+(260u^3-676u^2-260u-260)w^3+(-195u^4+432u^2+195)+(50u^5-130u^4+288u^3+288u^2-338u+130)w+(-169u^4-65u^2)=0$



Its resultant (here) is a polynomial of $u$ of thirty second degree with many approximately twenty five digital integer coefficients. It has many real roots, among them $0$ and $1$ and the largest of them is approximately $2.3242305780688903970$. For the resultant which is a polynomial of $w$ we have a similar situation: matrix, polynomial, and roots. Assuming that the function $f$ attains its minimum at a point $(u,w)$ which is a solution of system (*), it remains to check the values $f(u,w)$ for each pair of these non-negative roots. It turned out that $f(0,0)=f(1,1)=0$ and $f(u,w)>0.02$ for any other pair of the roots. Thus assuming that my Mathcad calculated the roots with an error at most $4cdot 10^{-9}$ , we have a proof.






share|cite|improve this answer











$endgroup$





















    0












    $begingroup$

    COMMENT.-Noting $A=sqrt{dfrac{13}{5}}$ we have after some easy calculation the equivalent inequality
    $$xyf(x)+Ayf(y)+A^2f(z)ge0$$ where $f(t)=dfrac{t^2-A^2}{t^2+1}$
    so $-A^2=-2,6le f(t)le0$ in the interval $[0,A]$ and $f(t)gt0$ otherwise increasing till $1$ as limit when $ttoinfty$.



    The fact that $f(A)=0$ makes easy the calculation with Lagrange multipliers: the minimum of the function
    $$F(x,y,z)= xyf(x)+Ayf(y)+A^2f(z)\text{ having }xyz=A^3text{ as restriction }$$
    This minimum is reached when $x=y=z=A$ and because of $F(A,A,A)=0$ we are done.



    ►However one wants to solve the problem with more basic mathematics. For example, by simple $AMge GM$ the form of $F(x,y,z)$ leads to the fact that the asked inequality is valid for all $(x,y,z)$ such that $sqrt[3]{f(x)f(y)f(z)}ge0$.



    Because of $xyz=A^3$ the positive numbers $x,y,z$ can not be all the three less than $A$ nor greater than $A$ so at least one of them should be less than $A$. If one of the other two is less than $A$ we are done. Therefore it has been proved this way the inequality when two of the three variables are less than $A$.



    ► Consequently it remains to finish, the case in which two among $x,y,z$ are greater than $A$ (say $f(x)gt0$ and $f(y)gt 0$ and $f(z)lt0$).



    We left this part for the OP and just displayed the above solution with Lagrange multipliers.






    share|cite|improve this answer









    $endgroup$













      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%2f2481332%2finequality-for-three-variables%23new-answer', 'question_page');
      }
      );

      Post as a guest















      Required, but never shown
























      2 Answers
      2






      active

      oldest

      votes








      2 Answers
      2






      active

      oldest

      votes









      active

      oldest

      votes






      active

      oldest

      votes









      1












      $begingroup$

      This answer is extremely technical and uses Mathcad calculations.



      By routine calculations the inequality can be simplified to



      $$frac{5}{18}left(frac xA +1+frac Ayright)ge frac xAcdot frac 1{x^2+1}+frac 1{y^2+1}+frac Aycdot frac 1{z^2+1},$$



      where $A=sqrtfrac{13}5$.



      Or, because $z^2=frac {A^6}{x^2y^2}$,



      $$frac{5}{18}left(frac xA +1+frac Ayright)ge frac xAcdot frac 1{x^2+1}+frac 1{y^2+1}+frac Aycdot frac { x^2y^2}{A^6 + x^2y^2},$$



      Substituting $u=frac xA$, $v=frac yA $ we obtain



      $$frac{5}{18}left(u+1+frac 1vright)ge ucdot frac 1{A^2u^2+1}+frac 1{A^2v^2+1}+frac 1vcdot frac {u^2v^2}{A^2 + u^2v^2}$$



      $$u+1+frac 1vge 18left(ucdot frac 1{13u^2+5}+frac 1{13v^2+5}+frac 1vcdot frac {u^2v^2}{13 + 5u^2v^2}right)$$



      Substituting $w=frac 1v$, we obtain



      $$u+1+wge 18left(frac u{13u^2+5}+frac {w^2}{13+5w^2}+ frac {wu^2}{13w^2 + 5u^2}right)$$



      $25(u^5w^2+u^2+w^5)+65(u^5-u^4w^3-u^4w^2+u^4+u^3w^4-u^3+u^2w^5-u^2w-uw^4-w^4+w^3+w^2)+144 (u^3w^2+u^2w^3+u^2w^2)-169(u^4w+ u^2w^4+uw^2)ge 0$



      The graphs suggest that this inequality is true. We may look for its proof as follows.



      Denote the left hand side of the last inequality by $f(u,w)$. We are going to show that $inf{f(u,w): u,wge 0}=0$. When one of the variables $u$ and $w$ is fixed, $f(u,w)$ becomes a polynomial with respect to the other (which we denote by $v$) with the leading coefficient at least $25$. So it attains its minimum when $v=0$ or $frac{partial f}{partial v}=0$. If $u=0$ then $f(u,w)=25w^5-65w^4+65w^3+65w^2=w^3(5w-8)^2+15w^4+w^3+65w^2ge 0$. If $w=0$ then $f(u,w)=65u^5+65u^4-65u^3+25u^2=u^2(8u-5)^2+65u^5+u^4+15u^3ge 0$.



      Conditions $frac{partial f}{partial u}=0$ and $frac{partial f}{partial w}=0$ yield the system (*)



      $130uw^5+(195u^2-338u-65)w^4+(-260u^3+288u)w^3+(125u^4-260u^3+432u^2+288u-169)w^2+(-676u^3-130u)w+
      325u^4+260u^3-195u^2+50u=0$



      $(325u^2+125)w^4+(260u^3-676u^2-260u-260)w^3+(-195u^4+432u^2+195)+(50u^5-130u^4+288u^3+288u^2-338u+130)w+(-169u^4-65u^2)=0$



      Its resultant (here) is a polynomial of $u$ of thirty second degree with many approximately twenty five digital integer coefficients. It has many real roots, among them $0$ and $1$ and the largest of them is approximately $2.3242305780688903970$. For the resultant which is a polynomial of $w$ we have a similar situation: matrix, polynomial, and roots. Assuming that the function $f$ attains its minimum at a point $(u,w)$ which is a solution of system (*), it remains to check the values $f(u,w)$ for each pair of these non-negative roots. It turned out that $f(0,0)=f(1,1)=0$ and $f(u,w)>0.02$ for any other pair of the roots. Thus assuming that my Mathcad calculated the roots with an error at most $4cdot 10^{-9}$ , we have a proof.






      share|cite|improve this answer











      $endgroup$


















        1












        $begingroup$

        This answer is extremely technical and uses Mathcad calculations.



        By routine calculations the inequality can be simplified to



        $$frac{5}{18}left(frac xA +1+frac Ayright)ge frac xAcdot frac 1{x^2+1}+frac 1{y^2+1}+frac Aycdot frac 1{z^2+1},$$



        where $A=sqrtfrac{13}5$.



        Or, because $z^2=frac {A^6}{x^2y^2}$,



        $$frac{5}{18}left(frac xA +1+frac Ayright)ge frac xAcdot frac 1{x^2+1}+frac 1{y^2+1}+frac Aycdot frac { x^2y^2}{A^6 + x^2y^2},$$



        Substituting $u=frac xA$, $v=frac yA $ we obtain



        $$frac{5}{18}left(u+1+frac 1vright)ge ucdot frac 1{A^2u^2+1}+frac 1{A^2v^2+1}+frac 1vcdot frac {u^2v^2}{A^2 + u^2v^2}$$



        $$u+1+frac 1vge 18left(ucdot frac 1{13u^2+5}+frac 1{13v^2+5}+frac 1vcdot frac {u^2v^2}{13 + 5u^2v^2}right)$$



        Substituting $w=frac 1v$, we obtain



        $$u+1+wge 18left(frac u{13u^2+5}+frac {w^2}{13+5w^2}+ frac {wu^2}{13w^2 + 5u^2}right)$$



        $25(u^5w^2+u^2+w^5)+65(u^5-u^4w^3-u^4w^2+u^4+u^3w^4-u^3+u^2w^5-u^2w-uw^4-w^4+w^3+w^2)+144 (u^3w^2+u^2w^3+u^2w^2)-169(u^4w+ u^2w^4+uw^2)ge 0$



        The graphs suggest that this inequality is true. We may look for its proof as follows.



        Denote the left hand side of the last inequality by $f(u,w)$. We are going to show that $inf{f(u,w): u,wge 0}=0$. When one of the variables $u$ and $w$ is fixed, $f(u,w)$ becomes a polynomial with respect to the other (which we denote by $v$) with the leading coefficient at least $25$. So it attains its minimum when $v=0$ or $frac{partial f}{partial v}=0$. If $u=0$ then $f(u,w)=25w^5-65w^4+65w^3+65w^2=w^3(5w-8)^2+15w^4+w^3+65w^2ge 0$. If $w=0$ then $f(u,w)=65u^5+65u^4-65u^3+25u^2=u^2(8u-5)^2+65u^5+u^4+15u^3ge 0$.



        Conditions $frac{partial f}{partial u}=0$ and $frac{partial f}{partial w}=0$ yield the system (*)



        $130uw^5+(195u^2-338u-65)w^4+(-260u^3+288u)w^3+(125u^4-260u^3+432u^2+288u-169)w^2+(-676u^3-130u)w+
        325u^4+260u^3-195u^2+50u=0$



        $(325u^2+125)w^4+(260u^3-676u^2-260u-260)w^3+(-195u^4+432u^2+195)+(50u^5-130u^4+288u^3+288u^2-338u+130)w+(-169u^4-65u^2)=0$



        Its resultant (here) is a polynomial of $u$ of thirty second degree with many approximately twenty five digital integer coefficients. It has many real roots, among them $0$ and $1$ and the largest of them is approximately $2.3242305780688903970$. For the resultant which is a polynomial of $w$ we have a similar situation: matrix, polynomial, and roots. Assuming that the function $f$ attains its minimum at a point $(u,w)$ which is a solution of system (*), it remains to check the values $f(u,w)$ for each pair of these non-negative roots. It turned out that $f(0,0)=f(1,1)=0$ and $f(u,w)>0.02$ for any other pair of the roots. Thus assuming that my Mathcad calculated the roots with an error at most $4cdot 10^{-9}$ , we have a proof.






        share|cite|improve this answer











        $endgroup$
















          1












          1








          1





          $begingroup$

          This answer is extremely technical and uses Mathcad calculations.



          By routine calculations the inequality can be simplified to



          $$frac{5}{18}left(frac xA +1+frac Ayright)ge frac xAcdot frac 1{x^2+1}+frac 1{y^2+1}+frac Aycdot frac 1{z^2+1},$$



          where $A=sqrtfrac{13}5$.



          Or, because $z^2=frac {A^6}{x^2y^2}$,



          $$frac{5}{18}left(frac xA +1+frac Ayright)ge frac xAcdot frac 1{x^2+1}+frac 1{y^2+1}+frac Aycdot frac { x^2y^2}{A^6 + x^2y^2},$$



          Substituting $u=frac xA$, $v=frac yA $ we obtain



          $$frac{5}{18}left(u+1+frac 1vright)ge ucdot frac 1{A^2u^2+1}+frac 1{A^2v^2+1}+frac 1vcdot frac {u^2v^2}{A^2 + u^2v^2}$$



          $$u+1+frac 1vge 18left(ucdot frac 1{13u^2+5}+frac 1{13v^2+5}+frac 1vcdot frac {u^2v^2}{13 + 5u^2v^2}right)$$



          Substituting $w=frac 1v$, we obtain



          $$u+1+wge 18left(frac u{13u^2+5}+frac {w^2}{13+5w^2}+ frac {wu^2}{13w^2 + 5u^2}right)$$



          $25(u^5w^2+u^2+w^5)+65(u^5-u^4w^3-u^4w^2+u^4+u^3w^4-u^3+u^2w^5-u^2w-uw^4-w^4+w^3+w^2)+144 (u^3w^2+u^2w^3+u^2w^2)-169(u^4w+ u^2w^4+uw^2)ge 0$



          The graphs suggest that this inequality is true. We may look for its proof as follows.



          Denote the left hand side of the last inequality by $f(u,w)$. We are going to show that $inf{f(u,w): u,wge 0}=0$. When one of the variables $u$ and $w$ is fixed, $f(u,w)$ becomes a polynomial with respect to the other (which we denote by $v$) with the leading coefficient at least $25$. So it attains its minimum when $v=0$ or $frac{partial f}{partial v}=0$. If $u=0$ then $f(u,w)=25w^5-65w^4+65w^3+65w^2=w^3(5w-8)^2+15w^4+w^3+65w^2ge 0$. If $w=0$ then $f(u,w)=65u^5+65u^4-65u^3+25u^2=u^2(8u-5)^2+65u^5+u^4+15u^3ge 0$.



          Conditions $frac{partial f}{partial u}=0$ and $frac{partial f}{partial w}=0$ yield the system (*)



          $130uw^5+(195u^2-338u-65)w^4+(-260u^3+288u)w^3+(125u^4-260u^3+432u^2+288u-169)w^2+(-676u^3-130u)w+
          325u^4+260u^3-195u^2+50u=0$



          $(325u^2+125)w^4+(260u^3-676u^2-260u-260)w^3+(-195u^4+432u^2+195)+(50u^5-130u^4+288u^3+288u^2-338u+130)w+(-169u^4-65u^2)=0$



          Its resultant (here) is a polynomial of $u$ of thirty second degree with many approximately twenty five digital integer coefficients. It has many real roots, among them $0$ and $1$ and the largest of them is approximately $2.3242305780688903970$. For the resultant which is a polynomial of $w$ we have a similar situation: matrix, polynomial, and roots. Assuming that the function $f$ attains its minimum at a point $(u,w)$ which is a solution of system (*), it remains to check the values $f(u,w)$ for each pair of these non-negative roots. It turned out that $f(0,0)=f(1,1)=0$ and $f(u,w)>0.02$ for any other pair of the roots. Thus assuming that my Mathcad calculated the roots with an error at most $4cdot 10^{-9}$ , we have a proof.






          share|cite|improve this answer











          $endgroup$



          This answer is extremely technical and uses Mathcad calculations.



          By routine calculations the inequality can be simplified to



          $$frac{5}{18}left(frac xA +1+frac Ayright)ge frac xAcdot frac 1{x^2+1}+frac 1{y^2+1}+frac Aycdot frac 1{z^2+1},$$



          where $A=sqrtfrac{13}5$.



          Or, because $z^2=frac {A^6}{x^2y^2}$,



          $$frac{5}{18}left(frac xA +1+frac Ayright)ge frac xAcdot frac 1{x^2+1}+frac 1{y^2+1}+frac Aycdot frac { x^2y^2}{A^6 + x^2y^2},$$



          Substituting $u=frac xA$, $v=frac yA $ we obtain



          $$frac{5}{18}left(u+1+frac 1vright)ge ucdot frac 1{A^2u^2+1}+frac 1{A^2v^2+1}+frac 1vcdot frac {u^2v^2}{A^2 + u^2v^2}$$



          $$u+1+frac 1vge 18left(ucdot frac 1{13u^2+5}+frac 1{13v^2+5}+frac 1vcdot frac {u^2v^2}{13 + 5u^2v^2}right)$$



          Substituting $w=frac 1v$, we obtain



          $$u+1+wge 18left(frac u{13u^2+5}+frac {w^2}{13+5w^2}+ frac {wu^2}{13w^2 + 5u^2}right)$$



          $25(u^5w^2+u^2+w^5)+65(u^5-u^4w^3-u^4w^2+u^4+u^3w^4-u^3+u^2w^5-u^2w-uw^4-w^4+w^3+w^2)+144 (u^3w^2+u^2w^3+u^2w^2)-169(u^4w+ u^2w^4+uw^2)ge 0$



          The graphs suggest that this inequality is true. We may look for its proof as follows.



          Denote the left hand side of the last inequality by $f(u,w)$. We are going to show that $inf{f(u,w): u,wge 0}=0$. When one of the variables $u$ and $w$ is fixed, $f(u,w)$ becomes a polynomial with respect to the other (which we denote by $v$) with the leading coefficient at least $25$. So it attains its minimum when $v=0$ or $frac{partial f}{partial v}=0$. If $u=0$ then $f(u,w)=25w^5-65w^4+65w^3+65w^2=w^3(5w-8)^2+15w^4+w^3+65w^2ge 0$. If $w=0$ then $f(u,w)=65u^5+65u^4-65u^3+25u^2=u^2(8u-5)^2+65u^5+u^4+15u^3ge 0$.



          Conditions $frac{partial f}{partial u}=0$ and $frac{partial f}{partial w}=0$ yield the system (*)



          $130uw^5+(195u^2-338u-65)w^4+(-260u^3+288u)w^3+(125u^4-260u^3+432u^2+288u-169)w^2+(-676u^3-130u)w+
          325u^4+260u^3-195u^2+50u=0$



          $(325u^2+125)w^4+(260u^3-676u^2-260u-260)w^3+(-195u^4+432u^2+195)+(50u^5-130u^4+288u^3+288u^2-338u+130)w+(-169u^4-65u^2)=0$



          Its resultant (here) is a polynomial of $u$ of thirty second degree with many approximately twenty five digital integer coefficients. It has many real roots, among them $0$ and $1$ and the largest of them is approximately $2.3242305780688903970$. For the resultant which is a polynomial of $w$ we have a similar situation: matrix, polynomial, and roots. Assuming that the function $f$ attains its minimum at a point $(u,w)$ which is a solution of system (*), it remains to check the values $f(u,w)$ for each pair of these non-negative roots. It turned out that $f(0,0)=f(1,1)=0$ and $f(u,w)>0.02$ for any other pair of the roots. Thus assuming that my Mathcad calculated the roots with an error at most $4cdot 10^{-9}$ , we have a proof.







          share|cite|improve this answer














          share|cite|improve this answer



          share|cite|improve this answer








          edited Oct 22 '17 at 11:32

























          answered Oct 21 '17 at 19:06









          Alex RavskyAlex Ravsky

          40.4k32282




          40.4k32282























              0












              $begingroup$

              COMMENT.-Noting $A=sqrt{dfrac{13}{5}}$ we have after some easy calculation the equivalent inequality
              $$xyf(x)+Ayf(y)+A^2f(z)ge0$$ where $f(t)=dfrac{t^2-A^2}{t^2+1}$
              so $-A^2=-2,6le f(t)le0$ in the interval $[0,A]$ and $f(t)gt0$ otherwise increasing till $1$ as limit when $ttoinfty$.



              The fact that $f(A)=0$ makes easy the calculation with Lagrange multipliers: the minimum of the function
              $$F(x,y,z)= xyf(x)+Ayf(y)+A^2f(z)\text{ having }xyz=A^3text{ as restriction }$$
              This minimum is reached when $x=y=z=A$ and because of $F(A,A,A)=0$ we are done.



              ►However one wants to solve the problem with more basic mathematics. For example, by simple $AMge GM$ the form of $F(x,y,z)$ leads to the fact that the asked inequality is valid for all $(x,y,z)$ such that $sqrt[3]{f(x)f(y)f(z)}ge0$.



              Because of $xyz=A^3$ the positive numbers $x,y,z$ can not be all the three less than $A$ nor greater than $A$ so at least one of them should be less than $A$. If one of the other two is less than $A$ we are done. Therefore it has been proved this way the inequality when two of the three variables are less than $A$.



              ► Consequently it remains to finish, the case in which two among $x,y,z$ are greater than $A$ (say $f(x)gt0$ and $f(y)gt 0$ and $f(z)lt0$).



              We left this part for the OP and just displayed the above solution with Lagrange multipliers.






              share|cite|improve this answer









              $endgroup$


















                0












                $begingroup$

                COMMENT.-Noting $A=sqrt{dfrac{13}{5}}$ we have after some easy calculation the equivalent inequality
                $$xyf(x)+Ayf(y)+A^2f(z)ge0$$ where $f(t)=dfrac{t^2-A^2}{t^2+1}$
                so $-A^2=-2,6le f(t)le0$ in the interval $[0,A]$ and $f(t)gt0$ otherwise increasing till $1$ as limit when $ttoinfty$.



                The fact that $f(A)=0$ makes easy the calculation with Lagrange multipliers: the minimum of the function
                $$F(x,y,z)= xyf(x)+Ayf(y)+A^2f(z)\text{ having }xyz=A^3text{ as restriction }$$
                This minimum is reached when $x=y=z=A$ and because of $F(A,A,A)=0$ we are done.



                ►However one wants to solve the problem with more basic mathematics. For example, by simple $AMge GM$ the form of $F(x,y,z)$ leads to the fact that the asked inequality is valid for all $(x,y,z)$ such that $sqrt[3]{f(x)f(y)f(z)}ge0$.



                Because of $xyz=A^3$ the positive numbers $x,y,z$ can not be all the three less than $A$ nor greater than $A$ so at least one of them should be less than $A$. If one of the other two is less than $A$ we are done. Therefore it has been proved this way the inequality when two of the three variables are less than $A$.



                ► Consequently it remains to finish, the case in which two among $x,y,z$ are greater than $A$ (say $f(x)gt0$ and $f(y)gt 0$ and $f(z)lt0$).



                We left this part for the OP and just displayed the above solution with Lagrange multipliers.






                share|cite|improve this answer









                $endgroup$
















                  0












                  0








                  0





                  $begingroup$

                  COMMENT.-Noting $A=sqrt{dfrac{13}{5}}$ we have after some easy calculation the equivalent inequality
                  $$xyf(x)+Ayf(y)+A^2f(z)ge0$$ where $f(t)=dfrac{t^2-A^2}{t^2+1}$
                  so $-A^2=-2,6le f(t)le0$ in the interval $[0,A]$ and $f(t)gt0$ otherwise increasing till $1$ as limit when $ttoinfty$.



                  The fact that $f(A)=0$ makes easy the calculation with Lagrange multipliers: the minimum of the function
                  $$F(x,y,z)= xyf(x)+Ayf(y)+A^2f(z)\text{ having }xyz=A^3text{ as restriction }$$
                  This minimum is reached when $x=y=z=A$ and because of $F(A,A,A)=0$ we are done.



                  ►However one wants to solve the problem with more basic mathematics. For example, by simple $AMge GM$ the form of $F(x,y,z)$ leads to the fact that the asked inequality is valid for all $(x,y,z)$ such that $sqrt[3]{f(x)f(y)f(z)}ge0$.



                  Because of $xyz=A^3$ the positive numbers $x,y,z$ can not be all the three less than $A$ nor greater than $A$ so at least one of them should be less than $A$. If one of the other two is less than $A$ we are done. Therefore it has been proved this way the inequality when two of the three variables are less than $A$.



                  ► Consequently it remains to finish, the case in which two among $x,y,z$ are greater than $A$ (say $f(x)gt0$ and $f(y)gt 0$ and $f(z)lt0$).



                  We left this part for the OP and just displayed the above solution with Lagrange multipliers.






                  share|cite|improve this answer









                  $endgroup$



                  COMMENT.-Noting $A=sqrt{dfrac{13}{5}}$ we have after some easy calculation the equivalent inequality
                  $$xyf(x)+Ayf(y)+A^2f(z)ge0$$ where $f(t)=dfrac{t^2-A^2}{t^2+1}$
                  so $-A^2=-2,6le f(t)le0$ in the interval $[0,A]$ and $f(t)gt0$ otherwise increasing till $1$ as limit when $ttoinfty$.



                  The fact that $f(A)=0$ makes easy the calculation with Lagrange multipliers: the minimum of the function
                  $$F(x,y,z)= xyf(x)+Ayf(y)+A^2f(z)\text{ having }xyz=A^3text{ as restriction }$$
                  This minimum is reached when $x=y=z=A$ and because of $F(A,A,A)=0$ we are done.



                  ►However one wants to solve the problem with more basic mathematics. For example, by simple $AMge GM$ the form of $F(x,y,z)$ leads to the fact that the asked inequality is valid for all $(x,y,z)$ such that $sqrt[3]{f(x)f(y)f(z)}ge0$.



                  Because of $xyz=A^3$ the positive numbers $x,y,z$ can not be all the three less than $A$ nor greater than $A$ so at least one of them should be less than $A$. If one of the other two is less than $A$ we are done. Therefore it has been proved this way the inequality when two of the three variables are less than $A$.



                  ► Consequently it remains to finish, the case in which two among $x,y,z$ are greater than $A$ (say $f(x)gt0$ and $f(y)gt 0$ and $f(z)lt0$).



                  We left this part for the OP and just displayed the above solution with Lagrange multipliers.







                  share|cite|improve this answer












                  share|cite|improve this answer



                  share|cite|improve this answer










                  answered Oct 23 '17 at 22:49









                  PiquitoPiquito

                  17.9k31438




                  17.9k31438






























                      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%2f2481332%2finequality-for-three-variables%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

                      Aardman Animations

                      Are they similar matrix

                      “minimization” problem in Euclidean space related to orthonormal basis