Finding a distribution with a given correlation to a Normal Distribution












0












$begingroup$


Below is my attempt to solve a problem which I believe has a solution. My work shows there is no solution. I believe I did something wrong. It could be related to the algebra.



Thanks,

Bob



Problem:

Let $X$ and $Y$ independent normally distributed random variables with a mean of $0$ and a standard deviation
of $1$. Let $K$ be a positive real number. Let $Z = Y + KX$ such that the correlation of $X$ and $Z$ is $frac{1}{2}$. Find $K$.

Answer:

If we select $K = 0$ then we get a correlation of $0$. If we select $K$ to be a very large number
then the correlation will be close to $1$.
begin{align*}
rho &= frac{1}{2} \
u_x &= 0 \
u_y &= 0 \
u_z &= u_y + K(u_y) = 0 + K(0) = 0 \
sigma_x &= 1 \
sigma_y &= 1 \
sigma_z^2 &= sigma_y^2 + K^2 sigma_x^2 + K(sigma_{xy}) \
sigma_{xy} &= 0 \
sigma_z^2 &= 1^2 + K^2 1^2 + K(0) = K^2 + 1 \
end{align*}

begin{align*}
sigma_{xz} &= E(XZ) - E(X)E(Z) \
E(X) &= 0 \
sigma_{xz} &= E(XZ) - (0)E(Z) = E(XZ) \
E(XZ) &= E( X(Y+KX)) = E(XY + KX^2) = E(XY) + KE(X^2) \
E(X^2) &= 1 \
end{align*}

Now we need to find $E(XY)$.
begin{align*}
E(XY) &= E(X) E(Y) = 0(0) \
E(XY) &= 0 \
E(XZ) &= 0 + K(1) = K \
sigma_{xz} &= K \
rho_{xz} &= frac{sigma_x sigma_z }{sigma_{xz}} \
frac{1}{2} &= frac{ sqrt{K^2+1} }{K} \
frac{K}{2} &= sqrt{K^2+1} \
frac{K^2}{4} &= K^2 + 1 \
frac{-3K^2}{4} &= 1 \
end{align*}

This equation has no real roots.










share|cite|improve this question









$endgroup$








  • 1




    $begingroup$
    Your correlation formula is inverse. Correlation is equal to covariance divided by product of standard deviations. Fix that and you get $K=1/sqrt{3}$ and $K=-1/sqrt{3}$
    $endgroup$
    – Sauhard Sharma
    Dec 23 '18 at 18:43


















0












$begingroup$


Below is my attempt to solve a problem which I believe has a solution. My work shows there is no solution. I believe I did something wrong. It could be related to the algebra.



Thanks,

Bob



Problem:

Let $X$ and $Y$ independent normally distributed random variables with a mean of $0$ and a standard deviation
of $1$. Let $K$ be a positive real number. Let $Z = Y + KX$ such that the correlation of $X$ and $Z$ is $frac{1}{2}$. Find $K$.

Answer:

If we select $K = 0$ then we get a correlation of $0$. If we select $K$ to be a very large number
then the correlation will be close to $1$.
begin{align*}
rho &= frac{1}{2} \
u_x &= 0 \
u_y &= 0 \
u_z &= u_y + K(u_y) = 0 + K(0) = 0 \
sigma_x &= 1 \
sigma_y &= 1 \
sigma_z^2 &= sigma_y^2 + K^2 sigma_x^2 + K(sigma_{xy}) \
sigma_{xy} &= 0 \
sigma_z^2 &= 1^2 + K^2 1^2 + K(0) = K^2 + 1 \
end{align*}

begin{align*}
sigma_{xz} &= E(XZ) - E(X)E(Z) \
E(X) &= 0 \
sigma_{xz} &= E(XZ) - (0)E(Z) = E(XZ) \
E(XZ) &= E( X(Y+KX)) = E(XY + KX^2) = E(XY) + KE(X^2) \
E(X^2) &= 1 \
end{align*}

Now we need to find $E(XY)$.
begin{align*}
E(XY) &= E(X) E(Y) = 0(0) \
E(XY) &= 0 \
E(XZ) &= 0 + K(1) = K \
sigma_{xz} &= K \
rho_{xz} &= frac{sigma_x sigma_z }{sigma_{xz}} \
frac{1}{2} &= frac{ sqrt{K^2+1} }{K} \
frac{K}{2} &= sqrt{K^2+1} \
frac{K^2}{4} &= K^2 + 1 \
frac{-3K^2}{4} &= 1 \
end{align*}

This equation has no real roots.










share|cite|improve this question









$endgroup$








  • 1




    $begingroup$
    Your correlation formula is inverse. Correlation is equal to covariance divided by product of standard deviations. Fix that and you get $K=1/sqrt{3}$ and $K=-1/sqrt{3}$
    $endgroup$
    – Sauhard Sharma
    Dec 23 '18 at 18:43
















0












0








0





$begingroup$


Below is my attempt to solve a problem which I believe has a solution. My work shows there is no solution. I believe I did something wrong. It could be related to the algebra.



Thanks,

Bob



Problem:

Let $X$ and $Y$ independent normally distributed random variables with a mean of $0$ and a standard deviation
of $1$. Let $K$ be a positive real number. Let $Z = Y + KX$ such that the correlation of $X$ and $Z$ is $frac{1}{2}$. Find $K$.

Answer:

If we select $K = 0$ then we get a correlation of $0$. If we select $K$ to be a very large number
then the correlation will be close to $1$.
begin{align*}
rho &= frac{1}{2} \
u_x &= 0 \
u_y &= 0 \
u_z &= u_y + K(u_y) = 0 + K(0) = 0 \
sigma_x &= 1 \
sigma_y &= 1 \
sigma_z^2 &= sigma_y^2 + K^2 sigma_x^2 + K(sigma_{xy}) \
sigma_{xy} &= 0 \
sigma_z^2 &= 1^2 + K^2 1^2 + K(0) = K^2 + 1 \
end{align*}

begin{align*}
sigma_{xz} &= E(XZ) - E(X)E(Z) \
E(X) &= 0 \
sigma_{xz} &= E(XZ) - (0)E(Z) = E(XZ) \
E(XZ) &= E( X(Y+KX)) = E(XY + KX^2) = E(XY) + KE(X^2) \
E(X^2) &= 1 \
end{align*}

Now we need to find $E(XY)$.
begin{align*}
E(XY) &= E(X) E(Y) = 0(0) \
E(XY) &= 0 \
E(XZ) &= 0 + K(1) = K \
sigma_{xz} &= K \
rho_{xz} &= frac{sigma_x sigma_z }{sigma_{xz}} \
frac{1}{2} &= frac{ sqrt{K^2+1} }{K} \
frac{K}{2} &= sqrt{K^2+1} \
frac{K^2}{4} &= K^2 + 1 \
frac{-3K^2}{4} &= 1 \
end{align*}

This equation has no real roots.










share|cite|improve this question









$endgroup$




Below is my attempt to solve a problem which I believe has a solution. My work shows there is no solution. I believe I did something wrong. It could be related to the algebra.



Thanks,

Bob



Problem:

Let $X$ and $Y$ independent normally distributed random variables with a mean of $0$ and a standard deviation
of $1$. Let $K$ be a positive real number. Let $Z = Y + KX$ such that the correlation of $X$ and $Z$ is $frac{1}{2}$. Find $K$.

Answer:

If we select $K = 0$ then we get a correlation of $0$. If we select $K$ to be a very large number
then the correlation will be close to $1$.
begin{align*}
rho &= frac{1}{2} \
u_x &= 0 \
u_y &= 0 \
u_z &= u_y + K(u_y) = 0 + K(0) = 0 \
sigma_x &= 1 \
sigma_y &= 1 \
sigma_z^2 &= sigma_y^2 + K^2 sigma_x^2 + K(sigma_{xy}) \
sigma_{xy} &= 0 \
sigma_z^2 &= 1^2 + K^2 1^2 + K(0) = K^2 + 1 \
end{align*}

begin{align*}
sigma_{xz} &= E(XZ) - E(X)E(Z) \
E(X) &= 0 \
sigma_{xz} &= E(XZ) - (0)E(Z) = E(XZ) \
E(XZ) &= E( X(Y+KX)) = E(XY + KX^2) = E(XY) + KE(X^2) \
E(X^2) &= 1 \
end{align*}

Now we need to find $E(XY)$.
begin{align*}
E(XY) &= E(X) E(Y) = 0(0) \
E(XY) &= 0 \
E(XZ) &= 0 + K(1) = K \
sigma_{xz} &= K \
rho_{xz} &= frac{sigma_x sigma_z }{sigma_{xz}} \
frac{1}{2} &= frac{ sqrt{K^2+1} }{K} \
frac{K}{2} &= sqrt{K^2+1} \
frac{K^2}{4} &= K^2 + 1 \
frac{-3K^2}{4} &= 1 \
end{align*}

This equation has no real roots.







probability






share|cite|improve this question













share|cite|improve this question











share|cite|improve this question




share|cite|improve this question










asked Dec 23 '18 at 18:19









BobBob

928515




928515








  • 1




    $begingroup$
    Your correlation formula is inverse. Correlation is equal to covariance divided by product of standard deviations. Fix that and you get $K=1/sqrt{3}$ and $K=-1/sqrt{3}$
    $endgroup$
    – Sauhard Sharma
    Dec 23 '18 at 18:43
















  • 1




    $begingroup$
    Your correlation formula is inverse. Correlation is equal to covariance divided by product of standard deviations. Fix that and you get $K=1/sqrt{3}$ and $K=-1/sqrt{3}$
    $endgroup$
    – Sauhard Sharma
    Dec 23 '18 at 18:43










1




1




$begingroup$
Your correlation formula is inverse. Correlation is equal to covariance divided by product of standard deviations. Fix that and you get $K=1/sqrt{3}$ and $K=-1/sqrt{3}$
$endgroup$
– Sauhard Sharma
Dec 23 '18 at 18:43






$begingroup$
Your correlation formula is inverse. Correlation is equal to covariance divided by product of standard deviations. Fix that and you get $K=1/sqrt{3}$ and $K=-1/sqrt{3}$
$endgroup$
– Sauhard Sharma
Dec 23 '18 at 18:43












2 Answers
2






active

oldest

votes


















0












$begingroup$

The formula for the correlation is incorrect (you mixed the numerator with the denominator). The corrected one is below (as already stated in the comments).



$$rho_{xz}=frac{sigma_{xz}}{sigma_xsigma_z}$$



It can help to check your assumptions in cases like this, for example your claim that 'if $K$ is very large the correlation approaches $1$' does not hold with $frac{sqrt{K^2+1}}{K}$.






share|cite|improve this answer









$endgroup$





















    0












    $begingroup$

    $Cov(X,Z)=E(XZ)=E(X(Y+kX))=kE(X^2)+E(X)E(Y)=k,Var(Z)=1+k^2.$ So we want $1/2=cor(X,Z)=frac{cov(X,Z)}{sqrt{var(X)var(Z)}}=k/sqrt{1+k^2}.$ So, $k=1/sqrt 3$ should work.






    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%2f3050584%2ffinding-a-distribution-with-a-given-correlation-to-a-normal-distribution%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









      0












      $begingroup$

      The formula for the correlation is incorrect (you mixed the numerator with the denominator). The corrected one is below (as already stated in the comments).



      $$rho_{xz}=frac{sigma_{xz}}{sigma_xsigma_z}$$



      It can help to check your assumptions in cases like this, for example your claim that 'if $K$ is very large the correlation approaches $1$' does not hold with $frac{sqrt{K^2+1}}{K}$.






      share|cite|improve this answer









      $endgroup$


















        0












        $begingroup$

        The formula for the correlation is incorrect (you mixed the numerator with the denominator). The corrected one is below (as already stated in the comments).



        $$rho_{xz}=frac{sigma_{xz}}{sigma_xsigma_z}$$



        It can help to check your assumptions in cases like this, for example your claim that 'if $K$ is very large the correlation approaches $1$' does not hold with $frac{sqrt{K^2+1}}{K}$.






        share|cite|improve this answer









        $endgroup$
















          0












          0








          0





          $begingroup$

          The formula for the correlation is incorrect (you mixed the numerator with the denominator). The corrected one is below (as already stated in the comments).



          $$rho_{xz}=frac{sigma_{xz}}{sigma_xsigma_z}$$



          It can help to check your assumptions in cases like this, for example your claim that 'if $K$ is very large the correlation approaches $1$' does not hold with $frac{sqrt{K^2+1}}{K}$.






          share|cite|improve this answer









          $endgroup$



          The formula for the correlation is incorrect (you mixed the numerator with the denominator). The corrected one is below (as already stated in the comments).



          $$rho_{xz}=frac{sigma_{xz}}{sigma_xsigma_z}$$



          It can help to check your assumptions in cases like this, for example your claim that 'if $K$ is very large the correlation approaches $1$' does not hold with $frac{sqrt{K^2+1}}{K}$.







          share|cite|improve this answer












          share|cite|improve this answer



          share|cite|improve this answer










          answered Dec 23 '18 at 18:52









          Cows quackCows quack

          1184




          1184























              0












              $begingroup$

              $Cov(X,Z)=E(XZ)=E(X(Y+kX))=kE(X^2)+E(X)E(Y)=k,Var(Z)=1+k^2.$ So we want $1/2=cor(X,Z)=frac{cov(X,Z)}{sqrt{var(X)var(Z)}}=k/sqrt{1+k^2}.$ So, $k=1/sqrt 3$ should work.






              share|cite|improve this answer









              $endgroup$


















                0












                $begingroup$

                $Cov(X,Z)=E(XZ)=E(X(Y+kX))=kE(X^2)+E(X)E(Y)=k,Var(Z)=1+k^2.$ So we want $1/2=cor(X,Z)=frac{cov(X,Z)}{sqrt{var(X)var(Z)}}=k/sqrt{1+k^2}.$ So, $k=1/sqrt 3$ should work.






                share|cite|improve this answer









                $endgroup$
















                  0












                  0








                  0





                  $begingroup$

                  $Cov(X,Z)=E(XZ)=E(X(Y+kX))=kE(X^2)+E(X)E(Y)=k,Var(Z)=1+k^2.$ So we want $1/2=cor(X,Z)=frac{cov(X,Z)}{sqrt{var(X)var(Z)}}=k/sqrt{1+k^2}.$ So, $k=1/sqrt 3$ should work.






                  share|cite|improve this answer









                  $endgroup$



                  $Cov(X,Z)=E(XZ)=E(X(Y+kX))=kE(X^2)+E(X)E(Y)=k,Var(Z)=1+k^2.$ So we want $1/2=cor(X,Z)=frac{cov(X,Z)}{sqrt{var(X)var(Z)}}=k/sqrt{1+k^2}.$ So, $k=1/sqrt 3$ should work.







                  share|cite|improve this answer












                  share|cite|improve this answer



                  share|cite|improve this answer










                  answered Dec 23 '18 at 18:55









                  John_WickJohn_Wick

                  1,616111




                  1,616111






























                      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%2f3050584%2ffinding-a-distribution-with-a-given-correlation-to-a-normal-distribution%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

                      Probability when a professor distributes a quiz and homework assignment to a class of n students.

                      Aardman Animations

                      Are they similar matrix