How to evaluate this nonelementary integral?












10












$begingroup$


Let $x>0$. I have to prove that



$$
int_{0}^{infty}frac{cos x}{x^p}dx=frac{pi}{2Gamma(p)cos(pfrac{pi}{2})}tag{1}
$$



by converting the integral on the left side to a double integral using the expression below:



$$
frac{1}{x^p}=frac{1}{Gamma(p)}int_{0}^{infty}e^{-xt}t^{p-1}dttag{2}
$$



By plugging $(2)$ into $(1)$ I get the following double integral:



$$
frac{1}{Gamma(p)}int_{0}^{infty}int_{0}^{infty}e^{-xt}t^{p-1}cos xdtdxtag{3}
$$



However, I unable to proceed any further as I am unclear as to what method should I use in order to compute this integral. I thought that an appropriate change of variables could transform it into a product of two gamma functions but I cannot see how that would work. Any help would be greatly appreciated.










share|cite|improve this question











$endgroup$








  • 1




    $begingroup$
    Change the order of integration, then use any method from Closed form for $ int_0^infty {frac{{{x^n}}}{{1 + {x^m}}}dx }$
    $endgroup$
    – projectilemotion
    Nov 24 '18 at 21:36


















10












$begingroup$


Let $x>0$. I have to prove that



$$
int_{0}^{infty}frac{cos x}{x^p}dx=frac{pi}{2Gamma(p)cos(pfrac{pi}{2})}tag{1}
$$



by converting the integral on the left side to a double integral using the expression below:



$$
frac{1}{x^p}=frac{1}{Gamma(p)}int_{0}^{infty}e^{-xt}t^{p-1}dttag{2}
$$



By plugging $(2)$ into $(1)$ I get the following double integral:



$$
frac{1}{Gamma(p)}int_{0}^{infty}int_{0}^{infty}e^{-xt}t^{p-1}cos xdtdxtag{3}
$$



However, I unable to proceed any further as I am unclear as to what method should I use in order to compute this integral. I thought that an appropriate change of variables could transform it into a product of two gamma functions but I cannot see how that would work. Any help would be greatly appreciated.










share|cite|improve this question











$endgroup$








  • 1




    $begingroup$
    Change the order of integration, then use any method from Closed form for $ int_0^infty {frac{{{x^n}}}{{1 + {x^m}}}dx }$
    $endgroup$
    – projectilemotion
    Nov 24 '18 at 21:36
















10












10








10


1



$begingroup$


Let $x>0$. I have to prove that



$$
int_{0}^{infty}frac{cos x}{x^p}dx=frac{pi}{2Gamma(p)cos(pfrac{pi}{2})}tag{1}
$$



by converting the integral on the left side to a double integral using the expression below:



$$
frac{1}{x^p}=frac{1}{Gamma(p)}int_{0}^{infty}e^{-xt}t^{p-1}dttag{2}
$$



By plugging $(2)$ into $(1)$ I get the following double integral:



$$
frac{1}{Gamma(p)}int_{0}^{infty}int_{0}^{infty}e^{-xt}t^{p-1}cos xdtdxtag{3}
$$



However, I unable to proceed any further as I am unclear as to what method should I use in order to compute this integral. I thought that an appropriate change of variables could transform it into a product of two gamma functions but I cannot see how that would work. Any help would be greatly appreciated.










share|cite|improve this question











$endgroup$




Let $x>0$. I have to prove that



$$
int_{0}^{infty}frac{cos x}{x^p}dx=frac{pi}{2Gamma(p)cos(pfrac{pi}{2})}tag{1}
$$



by converting the integral on the left side to a double integral using the expression below:



$$
frac{1}{x^p}=frac{1}{Gamma(p)}int_{0}^{infty}e^{-xt}t^{p-1}dttag{2}
$$



By plugging $(2)$ into $(1)$ I get the following double integral:



$$
frac{1}{Gamma(p)}int_{0}^{infty}int_{0}^{infty}e^{-xt}t^{p-1}cos xdtdxtag{3}
$$



However, I unable to proceed any further as I am unclear as to what method should I use in order to compute this integral. I thought that an appropriate change of variables could transform it into a product of two gamma functions but I cannot see how that would work. Any help would be greatly appreciated.







multivariable-calculus gamma-function






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Nov 24 '18 at 21:15









Key Flex

7,94461233




7,94461233










asked Nov 24 '18 at 21:14









PhillipPhillip

604




604








  • 1




    $begingroup$
    Change the order of integration, then use any method from Closed form for $ int_0^infty {frac{{{x^n}}}{{1 + {x^m}}}dx }$
    $endgroup$
    – projectilemotion
    Nov 24 '18 at 21:36
















  • 1




    $begingroup$
    Change the order of integration, then use any method from Closed form for $ int_0^infty {frac{{{x^n}}}{{1 + {x^m}}}dx }$
    $endgroup$
    – projectilemotion
    Nov 24 '18 at 21:36










1




1




$begingroup$
Change the order of integration, then use any method from Closed form for $ int_0^infty {frac{{{x^n}}}{{1 + {x^m}}}dx }$
$endgroup$
– projectilemotion
Nov 24 '18 at 21:36






$begingroup$
Change the order of integration, then use any method from Closed form for $ int_0^infty {frac{{{x^n}}}{{1 + {x^m}}}dx }$
$endgroup$
– projectilemotion
Nov 24 '18 at 21:36












4 Answers
4






active

oldest

votes


















9












$begingroup$

The Laplace transform of $cos x$ is $frac{s}{1+s^2}$ and the inverse Laplace transform of $frac{1}{x^p}$ is $frac{s^{p-1}}{Gamma(p)}$, hence
$$ int_{0}^{+infty}frac{cos x}{x^p},dx = frac{1}{Gamma(p)}int_{0}^{+infty}frac{s^p}{s^2+1},ds=frac{1}{Gamma(p)}int_{0}^{pi/2}left(tan uright)^p,du $$
equals
$$ begin{eqnarray*}frac{1}{Gamma(p)}int_{0}^{1} v^p (1-v^2)^{-(p+1)/2},dv&=&frac{1}{2,Gamma(p)}int_{0}^{1}w^{(p-1)/2}(1-w)^{-(p+1)/2},dw\& =& frac{Bleft(tfrac{1+p}{2},tfrac{1-p}{2}right)}{2,Gamma(p)}end{eqnarray*} $$
or
$$ frac{Gammaleft(frac{1+p}{2}right)Gammaleft(frac{1-p}{2}right)}{2,Gamma(p)}= frac{pi}{2,Gamma(p)sinleft(frac{pi}{2}(p+1)right)}=frac{pi}{2,Gamma(p)cosleft(frac{pi p}{2}right)}$$
as wanted. We have exploited the Beta function and the reflection formula for the $Gamma$ function.






share|cite|improve this answer









$endgroup$





















    5












    $begingroup$

    Your given integral is closely related to the Mellin transform and can be evaluated by using Ramanujan's Master Theorem.




    Ramanujan's Master Theorem



    Let $f(x)$ be an analytic function with a MacLaurin Expansion of the form
    $$f(x)=sum_{k=0}^{infty}frac{phi(k)}{k!}(-x)^k$$then the Mellin Transform of this function is given by
    $$int_0^{infty}x^{s-1}f(x)dx=Gamma(s)phi(-s)$$




    Therefore expand the cosine function as Taylor series expansion to get



    $$begin{align}
    mathfrak{I}=int_0^{infty}cos(x)x^{-p}dx&=int_0^{infty}x^{-p}sum_{n=0}^{infty}(-1)^nfrac{x^{2n}}{(2n)!}dx
    end{align}$$



    In order to bring the above integral in the wanted form for the usage of Ramanujan's Master Theorem apply the substitution $x^2=u$. So we further get



    $$begin{align}
    mathfrak{I}=int_0^{infty}x^{-p}sum_{n=0}^{infty}(-1)^nfrac{x^{2n}}{(2n)!}dx&=int_0^{infty}x^{-p}sum_{n=0}^{infty}frac{1}{(2n)!}(-x^2)^ndx\
    &=int_0^{infty}u^{-p/2}sum_{n=0}^{infty}frac{1}{(2n)!}(-u)^nfrac{du}{2sqrt{u}}\
    &=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{1}{(2n)!}(-u)^ndu\
    &=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{n!/(2n)!}{n!}(-u)^ndu
    end{align}$$



    By using the relation $Gamma(n)=(n-1)!$ which is valid for all $ninmathbb N$ we can consider the last integral as an application of Ramanujan's Master Theorem with $s=-frac{p-1}2$ and $phi(n)=frac{Gamma(n+1)}{Gamma(2n+1)}$. By finally using the Theorem we obtain



    $$begin{align}
    mathfrak{I}=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{n!/(2n)!}{n!}(-u)^ndu&=frac12Gammaleft(-frac{p-1}2right)frac{Gammaleft(frac{p-1}2+1right)}{Gammaleft(2left(frac{p-1}2right)+1right)}\
    &=frac1{2Gamma(p)}Gammaleft(1+frac{p-1}2right)Gammaleft(-frac{p-1}2right)
    end{align}$$



    Now by applying Euler's Reflection Formula with $z=1+frac{p-1}2$ we moreover get



    $$begin{align}
    mathfrak{I}=frac1{2Gamma(p)}Gammaleft(1+frac{p-1}2right)Gammaleft(-frac{p-1}2right)&=frac1{2Gamma(p)}frac{pi}{sinleft(pileft(1+frac{p-1}2right)right)}\
    &=frac1{2Gamma(p)}frac{pi}{sinleft(frac{ppi}2+frac{pi}2right)}\
    &=frac1{2Gamma(p)}frac{pi}{cosleft(frac{ppi}2right)}
    end{align}$$



    where within the last step the fundamental relation $sinleft(x+frac{pi}2right)=cos(x)$ was used. Thus for the original integral $mathfrak{I}$ we get




    $$mathfrak{I}=int_0^{infty}cos(x)x^{-p}dx=frac{pi}{2Gamma(p)cosleft(pfrac{pi}2right)}$$







    share|cite|improve this answer











    $endgroup$





















      3












      $begingroup$

      Hint: $displaystyleint_{0}^{infty}frac{cos x}{x^p}dx= text{Real part of}:int_{0}^{infty}frac{e^{iz}}{z^p}dz$ and use residue theorem. This has a pole of order $p$ hence the term $Gamma (p)$ in the denominator.






      share|cite|improve this answer









      $endgroup$













      • $begingroup$
        This only works for integral $p$, right?
        $endgroup$
        – AccidentalFourierTransform
        Nov 24 '18 at 22:13










      • $begingroup$
        @AccidentalFourierTransform: Your comment makes me wonder whether there is a way to bootstrap from integer $p$ to the complex $p$...
        $endgroup$
        – user21820
        Nov 25 '18 at 4:13










      • $begingroup$
        @user21820 Something in the spirit of the Identity theorem, or more generally Analytic continuation, I guess...
        $endgroup$
        – AccidentalFourierTransform
        Nov 25 '18 at 4:16










      • $begingroup$
        @AccidentalFourierTransform: But here the answer is a meromorphic function of $p$, and we don't even have an accumulation point. So do you have a real proof? =)
        $endgroup$
        – user21820
        Nov 25 '18 at 4:22










      • $begingroup$
        @user21820 Oh, no, not at all. I was just pointing out that Yidati's answer was only valid for integer $p$. I didn't want to suggest that I have a better answer.
        $endgroup$
        – AccidentalFourierTransform
        Nov 25 '18 at 4:26



















      0












      $begingroup$

      So let us follow your initial line of thought and convert the integral to a double integral. As you correctly observe, as
      $$frac{1}{x^p} = frac{1}{Gamma (p)} int_0^infty e^{-xt} t^{p - 1} , dt,$$
      which, by the way, is just the Laplace transform for the function $x^{p -1}$, as a double integral your integral can be rewritten as
      $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty int_0^infty e^{-xt} cos x t^{p - 1} , dt , dx,$$
      or
      $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty t^{p - 1} int_0^infty e^{-xt} cos x , dx , dt,$$
      after changing the order of integration.



      The inner $x$-integral can be readily found. Either using integration by parts twice, or recognising the integral as the Laplace transform for the function $cos x$, as
      $$int_0^infty e^{-xt} cos x , dx = frac{t}{1 + t^2},$$
      we have
      $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty frac{t^p}{1 + t^2} , dt.$$
      Enforcing a substitution of $t mapsto sqrt{t}$ leads to
      begin{align}
      int_0^infty frac{cos x}{x^p} , dx &= frac{1}{2 Gamma (p)} int_0^infty frac{t^{frac{p}{2} - frac{1}{2}}}{1 + t} , dt = frac{1}{2 Gamma (p)} int_0^infty frac{t^{frac{p + 1}{2} - 1}}{(1 + t)^{frac{p + 1}{2} + frac{1 - p}{2}}}.
      end{align}

      As this is exactly of the form of the Beta function (see the second of the integral representations in the link) we have
      begin{align}
      int_0^infty frac{cos x}{x^p} , dx &= frac{1}{2 Gamma (p)} text{B} left (frac{p + 1}{2}, frac{1 - p}{2} right )\
      &= frac{1}{2 Gamma (p)} Gamma left (frac{p}{2} + frac{1}{2} right ) Gamma left (frac{1}{2} - frac{p}{2} right ) \
      &= frac{1}{2 Gamma (p)} Gamma left (frac{p}{2} + frac{1}{2} right ) Gamma left [1 - left (frac{p}{2} + frac{1}{2} right ) right ] \
      &= frac{1}{2 Gamma (p)} frac{pi}{sin (p + 1)pi/2} qquad (*)\
      &= frac{pi}{2 Gamma (p) cos left (frac{pi p}{2} right )},
      end{align}

      as required. Note Euler's reflection formula was used in ($*$).






      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%2f3012090%2fhow-to-evaluate-this-nonelementary-integral%23new-answer', 'question_page');
        }
        );

        Post as a guest















        Required, but never shown

























        4 Answers
        4






        active

        oldest

        votes








        4 Answers
        4






        active

        oldest

        votes









        active

        oldest

        votes






        active

        oldest

        votes









        9












        $begingroup$

        The Laplace transform of $cos x$ is $frac{s}{1+s^2}$ and the inverse Laplace transform of $frac{1}{x^p}$ is $frac{s^{p-1}}{Gamma(p)}$, hence
        $$ int_{0}^{+infty}frac{cos x}{x^p},dx = frac{1}{Gamma(p)}int_{0}^{+infty}frac{s^p}{s^2+1},ds=frac{1}{Gamma(p)}int_{0}^{pi/2}left(tan uright)^p,du $$
        equals
        $$ begin{eqnarray*}frac{1}{Gamma(p)}int_{0}^{1} v^p (1-v^2)^{-(p+1)/2},dv&=&frac{1}{2,Gamma(p)}int_{0}^{1}w^{(p-1)/2}(1-w)^{-(p+1)/2},dw\& =& frac{Bleft(tfrac{1+p}{2},tfrac{1-p}{2}right)}{2,Gamma(p)}end{eqnarray*} $$
        or
        $$ frac{Gammaleft(frac{1+p}{2}right)Gammaleft(frac{1-p}{2}right)}{2,Gamma(p)}= frac{pi}{2,Gamma(p)sinleft(frac{pi}{2}(p+1)right)}=frac{pi}{2,Gamma(p)cosleft(frac{pi p}{2}right)}$$
        as wanted. We have exploited the Beta function and the reflection formula for the $Gamma$ function.






        share|cite|improve this answer









        $endgroup$


















          9












          $begingroup$

          The Laplace transform of $cos x$ is $frac{s}{1+s^2}$ and the inverse Laplace transform of $frac{1}{x^p}$ is $frac{s^{p-1}}{Gamma(p)}$, hence
          $$ int_{0}^{+infty}frac{cos x}{x^p},dx = frac{1}{Gamma(p)}int_{0}^{+infty}frac{s^p}{s^2+1},ds=frac{1}{Gamma(p)}int_{0}^{pi/2}left(tan uright)^p,du $$
          equals
          $$ begin{eqnarray*}frac{1}{Gamma(p)}int_{0}^{1} v^p (1-v^2)^{-(p+1)/2},dv&=&frac{1}{2,Gamma(p)}int_{0}^{1}w^{(p-1)/2}(1-w)^{-(p+1)/2},dw\& =& frac{Bleft(tfrac{1+p}{2},tfrac{1-p}{2}right)}{2,Gamma(p)}end{eqnarray*} $$
          or
          $$ frac{Gammaleft(frac{1+p}{2}right)Gammaleft(frac{1-p}{2}right)}{2,Gamma(p)}= frac{pi}{2,Gamma(p)sinleft(frac{pi}{2}(p+1)right)}=frac{pi}{2,Gamma(p)cosleft(frac{pi p}{2}right)}$$
          as wanted. We have exploited the Beta function and the reflection formula for the $Gamma$ function.






          share|cite|improve this answer









          $endgroup$
















            9












            9








            9





            $begingroup$

            The Laplace transform of $cos x$ is $frac{s}{1+s^2}$ and the inverse Laplace transform of $frac{1}{x^p}$ is $frac{s^{p-1}}{Gamma(p)}$, hence
            $$ int_{0}^{+infty}frac{cos x}{x^p},dx = frac{1}{Gamma(p)}int_{0}^{+infty}frac{s^p}{s^2+1},ds=frac{1}{Gamma(p)}int_{0}^{pi/2}left(tan uright)^p,du $$
            equals
            $$ begin{eqnarray*}frac{1}{Gamma(p)}int_{0}^{1} v^p (1-v^2)^{-(p+1)/2},dv&=&frac{1}{2,Gamma(p)}int_{0}^{1}w^{(p-1)/2}(1-w)^{-(p+1)/2},dw\& =& frac{Bleft(tfrac{1+p}{2},tfrac{1-p}{2}right)}{2,Gamma(p)}end{eqnarray*} $$
            or
            $$ frac{Gammaleft(frac{1+p}{2}right)Gammaleft(frac{1-p}{2}right)}{2,Gamma(p)}= frac{pi}{2,Gamma(p)sinleft(frac{pi}{2}(p+1)right)}=frac{pi}{2,Gamma(p)cosleft(frac{pi p}{2}right)}$$
            as wanted. We have exploited the Beta function and the reflection formula for the $Gamma$ function.






            share|cite|improve this answer









            $endgroup$



            The Laplace transform of $cos x$ is $frac{s}{1+s^2}$ and the inverse Laplace transform of $frac{1}{x^p}$ is $frac{s^{p-1}}{Gamma(p)}$, hence
            $$ int_{0}^{+infty}frac{cos x}{x^p},dx = frac{1}{Gamma(p)}int_{0}^{+infty}frac{s^p}{s^2+1},ds=frac{1}{Gamma(p)}int_{0}^{pi/2}left(tan uright)^p,du $$
            equals
            $$ begin{eqnarray*}frac{1}{Gamma(p)}int_{0}^{1} v^p (1-v^2)^{-(p+1)/2},dv&=&frac{1}{2,Gamma(p)}int_{0}^{1}w^{(p-1)/2}(1-w)^{-(p+1)/2},dw\& =& frac{Bleft(tfrac{1+p}{2},tfrac{1-p}{2}right)}{2,Gamma(p)}end{eqnarray*} $$
            or
            $$ frac{Gammaleft(frac{1+p}{2}right)Gammaleft(frac{1-p}{2}right)}{2,Gamma(p)}= frac{pi}{2,Gamma(p)sinleft(frac{pi}{2}(p+1)right)}=frac{pi}{2,Gamma(p)cosleft(frac{pi p}{2}right)}$$
            as wanted. We have exploited the Beta function and the reflection formula for the $Gamma$ function.







            share|cite|improve this answer












            share|cite|improve this answer



            share|cite|improve this answer










            answered Nov 24 '18 at 21:27









            Jack D'AurizioJack D'Aurizio

            1




            1























                5












                $begingroup$

                Your given integral is closely related to the Mellin transform and can be evaluated by using Ramanujan's Master Theorem.




                Ramanujan's Master Theorem



                Let $f(x)$ be an analytic function with a MacLaurin Expansion of the form
                $$f(x)=sum_{k=0}^{infty}frac{phi(k)}{k!}(-x)^k$$then the Mellin Transform of this function is given by
                $$int_0^{infty}x^{s-1}f(x)dx=Gamma(s)phi(-s)$$




                Therefore expand the cosine function as Taylor series expansion to get



                $$begin{align}
                mathfrak{I}=int_0^{infty}cos(x)x^{-p}dx&=int_0^{infty}x^{-p}sum_{n=0}^{infty}(-1)^nfrac{x^{2n}}{(2n)!}dx
                end{align}$$



                In order to bring the above integral in the wanted form for the usage of Ramanujan's Master Theorem apply the substitution $x^2=u$. So we further get



                $$begin{align}
                mathfrak{I}=int_0^{infty}x^{-p}sum_{n=0}^{infty}(-1)^nfrac{x^{2n}}{(2n)!}dx&=int_0^{infty}x^{-p}sum_{n=0}^{infty}frac{1}{(2n)!}(-x^2)^ndx\
                &=int_0^{infty}u^{-p/2}sum_{n=0}^{infty}frac{1}{(2n)!}(-u)^nfrac{du}{2sqrt{u}}\
                &=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{1}{(2n)!}(-u)^ndu\
                &=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{n!/(2n)!}{n!}(-u)^ndu
                end{align}$$



                By using the relation $Gamma(n)=(n-1)!$ which is valid for all $ninmathbb N$ we can consider the last integral as an application of Ramanujan's Master Theorem with $s=-frac{p-1}2$ and $phi(n)=frac{Gamma(n+1)}{Gamma(2n+1)}$. By finally using the Theorem we obtain



                $$begin{align}
                mathfrak{I}=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{n!/(2n)!}{n!}(-u)^ndu&=frac12Gammaleft(-frac{p-1}2right)frac{Gammaleft(frac{p-1}2+1right)}{Gammaleft(2left(frac{p-1}2right)+1right)}\
                &=frac1{2Gamma(p)}Gammaleft(1+frac{p-1}2right)Gammaleft(-frac{p-1}2right)
                end{align}$$



                Now by applying Euler's Reflection Formula with $z=1+frac{p-1}2$ we moreover get



                $$begin{align}
                mathfrak{I}=frac1{2Gamma(p)}Gammaleft(1+frac{p-1}2right)Gammaleft(-frac{p-1}2right)&=frac1{2Gamma(p)}frac{pi}{sinleft(pileft(1+frac{p-1}2right)right)}\
                &=frac1{2Gamma(p)}frac{pi}{sinleft(frac{ppi}2+frac{pi}2right)}\
                &=frac1{2Gamma(p)}frac{pi}{cosleft(frac{ppi}2right)}
                end{align}$$



                where within the last step the fundamental relation $sinleft(x+frac{pi}2right)=cos(x)$ was used. Thus for the original integral $mathfrak{I}$ we get




                $$mathfrak{I}=int_0^{infty}cos(x)x^{-p}dx=frac{pi}{2Gamma(p)cosleft(pfrac{pi}2right)}$$







                share|cite|improve this answer











                $endgroup$


















                  5












                  $begingroup$

                  Your given integral is closely related to the Mellin transform and can be evaluated by using Ramanujan's Master Theorem.




                  Ramanujan's Master Theorem



                  Let $f(x)$ be an analytic function with a MacLaurin Expansion of the form
                  $$f(x)=sum_{k=0}^{infty}frac{phi(k)}{k!}(-x)^k$$then the Mellin Transform of this function is given by
                  $$int_0^{infty}x^{s-1}f(x)dx=Gamma(s)phi(-s)$$




                  Therefore expand the cosine function as Taylor series expansion to get



                  $$begin{align}
                  mathfrak{I}=int_0^{infty}cos(x)x^{-p}dx&=int_0^{infty}x^{-p}sum_{n=0}^{infty}(-1)^nfrac{x^{2n}}{(2n)!}dx
                  end{align}$$



                  In order to bring the above integral in the wanted form for the usage of Ramanujan's Master Theorem apply the substitution $x^2=u$. So we further get



                  $$begin{align}
                  mathfrak{I}=int_0^{infty}x^{-p}sum_{n=0}^{infty}(-1)^nfrac{x^{2n}}{(2n)!}dx&=int_0^{infty}x^{-p}sum_{n=0}^{infty}frac{1}{(2n)!}(-x^2)^ndx\
                  &=int_0^{infty}u^{-p/2}sum_{n=0}^{infty}frac{1}{(2n)!}(-u)^nfrac{du}{2sqrt{u}}\
                  &=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{1}{(2n)!}(-u)^ndu\
                  &=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{n!/(2n)!}{n!}(-u)^ndu
                  end{align}$$



                  By using the relation $Gamma(n)=(n-1)!$ which is valid for all $ninmathbb N$ we can consider the last integral as an application of Ramanujan's Master Theorem with $s=-frac{p-1}2$ and $phi(n)=frac{Gamma(n+1)}{Gamma(2n+1)}$. By finally using the Theorem we obtain



                  $$begin{align}
                  mathfrak{I}=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{n!/(2n)!}{n!}(-u)^ndu&=frac12Gammaleft(-frac{p-1}2right)frac{Gammaleft(frac{p-1}2+1right)}{Gammaleft(2left(frac{p-1}2right)+1right)}\
                  &=frac1{2Gamma(p)}Gammaleft(1+frac{p-1}2right)Gammaleft(-frac{p-1}2right)
                  end{align}$$



                  Now by applying Euler's Reflection Formula with $z=1+frac{p-1}2$ we moreover get



                  $$begin{align}
                  mathfrak{I}=frac1{2Gamma(p)}Gammaleft(1+frac{p-1}2right)Gammaleft(-frac{p-1}2right)&=frac1{2Gamma(p)}frac{pi}{sinleft(pileft(1+frac{p-1}2right)right)}\
                  &=frac1{2Gamma(p)}frac{pi}{sinleft(frac{ppi}2+frac{pi}2right)}\
                  &=frac1{2Gamma(p)}frac{pi}{cosleft(frac{ppi}2right)}
                  end{align}$$



                  where within the last step the fundamental relation $sinleft(x+frac{pi}2right)=cos(x)$ was used. Thus for the original integral $mathfrak{I}$ we get




                  $$mathfrak{I}=int_0^{infty}cos(x)x^{-p}dx=frac{pi}{2Gamma(p)cosleft(pfrac{pi}2right)}$$







                  share|cite|improve this answer











                  $endgroup$
















                    5












                    5








                    5





                    $begingroup$

                    Your given integral is closely related to the Mellin transform and can be evaluated by using Ramanujan's Master Theorem.




                    Ramanujan's Master Theorem



                    Let $f(x)$ be an analytic function with a MacLaurin Expansion of the form
                    $$f(x)=sum_{k=0}^{infty}frac{phi(k)}{k!}(-x)^k$$then the Mellin Transform of this function is given by
                    $$int_0^{infty}x^{s-1}f(x)dx=Gamma(s)phi(-s)$$




                    Therefore expand the cosine function as Taylor series expansion to get



                    $$begin{align}
                    mathfrak{I}=int_0^{infty}cos(x)x^{-p}dx&=int_0^{infty}x^{-p}sum_{n=0}^{infty}(-1)^nfrac{x^{2n}}{(2n)!}dx
                    end{align}$$



                    In order to bring the above integral in the wanted form for the usage of Ramanujan's Master Theorem apply the substitution $x^2=u$. So we further get



                    $$begin{align}
                    mathfrak{I}=int_0^{infty}x^{-p}sum_{n=0}^{infty}(-1)^nfrac{x^{2n}}{(2n)!}dx&=int_0^{infty}x^{-p}sum_{n=0}^{infty}frac{1}{(2n)!}(-x^2)^ndx\
                    &=int_0^{infty}u^{-p/2}sum_{n=0}^{infty}frac{1}{(2n)!}(-u)^nfrac{du}{2sqrt{u}}\
                    &=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{1}{(2n)!}(-u)^ndu\
                    &=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{n!/(2n)!}{n!}(-u)^ndu
                    end{align}$$



                    By using the relation $Gamma(n)=(n-1)!$ which is valid for all $ninmathbb N$ we can consider the last integral as an application of Ramanujan's Master Theorem with $s=-frac{p-1}2$ and $phi(n)=frac{Gamma(n+1)}{Gamma(2n+1)}$. By finally using the Theorem we obtain



                    $$begin{align}
                    mathfrak{I}=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{n!/(2n)!}{n!}(-u)^ndu&=frac12Gammaleft(-frac{p-1}2right)frac{Gammaleft(frac{p-1}2+1right)}{Gammaleft(2left(frac{p-1}2right)+1right)}\
                    &=frac1{2Gamma(p)}Gammaleft(1+frac{p-1}2right)Gammaleft(-frac{p-1}2right)
                    end{align}$$



                    Now by applying Euler's Reflection Formula with $z=1+frac{p-1}2$ we moreover get



                    $$begin{align}
                    mathfrak{I}=frac1{2Gamma(p)}Gammaleft(1+frac{p-1}2right)Gammaleft(-frac{p-1}2right)&=frac1{2Gamma(p)}frac{pi}{sinleft(pileft(1+frac{p-1}2right)right)}\
                    &=frac1{2Gamma(p)}frac{pi}{sinleft(frac{ppi}2+frac{pi}2right)}\
                    &=frac1{2Gamma(p)}frac{pi}{cosleft(frac{ppi}2right)}
                    end{align}$$



                    where within the last step the fundamental relation $sinleft(x+frac{pi}2right)=cos(x)$ was used. Thus for the original integral $mathfrak{I}$ we get




                    $$mathfrak{I}=int_0^{infty}cos(x)x^{-p}dx=frac{pi}{2Gamma(p)cosleft(pfrac{pi}2right)}$$







                    share|cite|improve this answer











                    $endgroup$



                    Your given integral is closely related to the Mellin transform and can be evaluated by using Ramanujan's Master Theorem.




                    Ramanujan's Master Theorem



                    Let $f(x)$ be an analytic function with a MacLaurin Expansion of the form
                    $$f(x)=sum_{k=0}^{infty}frac{phi(k)}{k!}(-x)^k$$then the Mellin Transform of this function is given by
                    $$int_0^{infty}x^{s-1}f(x)dx=Gamma(s)phi(-s)$$




                    Therefore expand the cosine function as Taylor series expansion to get



                    $$begin{align}
                    mathfrak{I}=int_0^{infty}cos(x)x^{-p}dx&=int_0^{infty}x^{-p}sum_{n=0}^{infty}(-1)^nfrac{x^{2n}}{(2n)!}dx
                    end{align}$$



                    In order to bring the above integral in the wanted form for the usage of Ramanujan's Master Theorem apply the substitution $x^2=u$. So we further get



                    $$begin{align}
                    mathfrak{I}=int_0^{infty}x^{-p}sum_{n=0}^{infty}(-1)^nfrac{x^{2n}}{(2n)!}dx&=int_0^{infty}x^{-p}sum_{n=0}^{infty}frac{1}{(2n)!}(-x^2)^ndx\
                    &=int_0^{infty}u^{-p/2}sum_{n=0}^{infty}frac{1}{(2n)!}(-u)^nfrac{du}{2sqrt{u}}\
                    &=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{1}{(2n)!}(-u)^ndu\
                    &=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{n!/(2n)!}{n!}(-u)^ndu
                    end{align}$$



                    By using the relation $Gamma(n)=(n-1)!$ which is valid for all $ninmathbb N$ we can consider the last integral as an application of Ramanujan's Master Theorem with $s=-frac{p-1}2$ and $phi(n)=frac{Gamma(n+1)}{Gamma(2n+1)}$. By finally using the Theorem we obtain



                    $$begin{align}
                    mathfrak{I}=frac12int_0^{infty}u^{-(p+1)/2}sum_{n=0}^{infty}frac{n!/(2n)!}{n!}(-u)^ndu&=frac12Gammaleft(-frac{p-1}2right)frac{Gammaleft(frac{p-1}2+1right)}{Gammaleft(2left(frac{p-1}2right)+1right)}\
                    &=frac1{2Gamma(p)}Gammaleft(1+frac{p-1}2right)Gammaleft(-frac{p-1}2right)
                    end{align}$$



                    Now by applying Euler's Reflection Formula with $z=1+frac{p-1}2$ we moreover get



                    $$begin{align}
                    mathfrak{I}=frac1{2Gamma(p)}Gammaleft(1+frac{p-1}2right)Gammaleft(-frac{p-1}2right)&=frac1{2Gamma(p)}frac{pi}{sinleft(pileft(1+frac{p-1}2right)right)}\
                    &=frac1{2Gamma(p)}frac{pi}{sinleft(frac{ppi}2+frac{pi}2right)}\
                    &=frac1{2Gamma(p)}frac{pi}{cosleft(frac{ppi}2right)}
                    end{align}$$



                    where within the last step the fundamental relation $sinleft(x+frac{pi}2right)=cos(x)$ was used. Thus for the original integral $mathfrak{I}$ we get




                    $$mathfrak{I}=int_0^{infty}cos(x)x^{-p}dx=frac{pi}{2Gamma(p)cosleft(pfrac{pi}2right)}$$








                    share|cite|improve this answer














                    share|cite|improve this answer



                    share|cite|improve this answer








                    edited Jan 8 at 10:21

























                    answered Nov 24 '18 at 21:42









                    mrtaurhomrtaurho

                    4,39621235




                    4,39621235























                        3












                        $begingroup$

                        Hint: $displaystyleint_{0}^{infty}frac{cos x}{x^p}dx= text{Real part of}:int_{0}^{infty}frac{e^{iz}}{z^p}dz$ and use residue theorem. This has a pole of order $p$ hence the term $Gamma (p)$ in the denominator.






                        share|cite|improve this answer









                        $endgroup$













                        • $begingroup$
                          This only works for integral $p$, right?
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 24 '18 at 22:13










                        • $begingroup$
                          @AccidentalFourierTransform: Your comment makes me wonder whether there is a way to bootstrap from integer $p$ to the complex $p$...
                          $endgroup$
                          – user21820
                          Nov 25 '18 at 4:13










                        • $begingroup$
                          @user21820 Something in the spirit of the Identity theorem, or more generally Analytic continuation, I guess...
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 25 '18 at 4:16










                        • $begingroup$
                          @AccidentalFourierTransform: But here the answer is a meromorphic function of $p$, and we don't even have an accumulation point. So do you have a real proof? =)
                          $endgroup$
                          – user21820
                          Nov 25 '18 at 4:22










                        • $begingroup$
                          @user21820 Oh, no, not at all. I was just pointing out that Yidati's answer was only valid for integer $p$. I didn't want to suggest that I have a better answer.
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 25 '18 at 4:26
















                        3












                        $begingroup$

                        Hint: $displaystyleint_{0}^{infty}frac{cos x}{x^p}dx= text{Real part of}:int_{0}^{infty}frac{e^{iz}}{z^p}dz$ and use residue theorem. This has a pole of order $p$ hence the term $Gamma (p)$ in the denominator.






                        share|cite|improve this answer









                        $endgroup$













                        • $begingroup$
                          This only works for integral $p$, right?
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 24 '18 at 22:13










                        • $begingroup$
                          @AccidentalFourierTransform: Your comment makes me wonder whether there is a way to bootstrap from integer $p$ to the complex $p$...
                          $endgroup$
                          – user21820
                          Nov 25 '18 at 4:13










                        • $begingroup$
                          @user21820 Something in the spirit of the Identity theorem, or more generally Analytic continuation, I guess...
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 25 '18 at 4:16










                        • $begingroup$
                          @AccidentalFourierTransform: But here the answer is a meromorphic function of $p$, and we don't even have an accumulation point. So do you have a real proof? =)
                          $endgroup$
                          – user21820
                          Nov 25 '18 at 4:22










                        • $begingroup$
                          @user21820 Oh, no, not at all. I was just pointing out that Yidati's answer was only valid for integer $p$. I didn't want to suggest that I have a better answer.
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 25 '18 at 4:26














                        3












                        3








                        3





                        $begingroup$

                        Hint: $displaystyleint_{0}^{infty}frac{cos x}{x^p}dx= text{Real part of}:int_{0}^{infty}frac{e^{iz}}{z^p}dz$ and use residue theorem. This has a pole of order $p$ hence the term $Gamma (p)$ in the denominator.






                        share|cite|improve this answer









                        $endgroup$



                        Hint: $displaystyleint_{0}^{infty}frac{cos x}{x^p}dx= text{Real part of}:int_{0}^{infty}frac{e^{iz}}{z^p}dz$ and use residue theorem. This has a pole of order $p$ hence the term $Gamma (p)$ in the denominator.







                        share|cite|improve this answer












                        share|cite|improve this answer



                        share|cite|improve this answer










                        answered Nov 24 '18 at 21:22









                        Yadati KiranYadati Kiran

                        1,773619




                        1,773619












                        • $begingroup$
                          This only works for integral $p$, right?
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 24 '18 at 22:13










                        • $begingroup$
                          @AccidentalFourierTransform: Your comment makes me wonder whether there is a way to bootstrap from integer $p$ to the complex $p$...
                          $endgroup$
                          – user21820
                          Nov 25 '18 at 4:13










                        • $begingroup$
                          @user21820 Something in the spirit of the Identity theorem, or more generally Analytic continuation, I guess...
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 25 '18 at 4:16










                        • $begingroup$
                          @AccidentalFourierTransform: But here the answer is a meromorphic function of $p$, and we don't even have an accumulation point. So do you have a real proof? =)
                          $endgroup$
                          – user21820
                          Nov 25 '18 at 4:22










                        • $begingroup$
                          @user21820 Oh, no, not at all. I was just pointing out that Yidati's answer was only valid for integer $p$. I didn't want to suggest that I have a better answer.
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 25 '18 at 4:26


















                        • $begingroup$
                          This only works for integral $p$, right?
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 24 '18 at 22:13










                        • $begingroup$
                          @AccidentalFourierTransform: Your comment makes me wonder whether there is a way to bootstrap from integer $p$ to the complex $p$...
                          $endgroup$
                          – user21820
                          Nov 25 '18 at 4:13










                        • $begingroup$
                          @user21820 Something in the spirit of the Identity theorem, or more generally Analytic continuation, I guess...
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 25 '18 at 4:16










                        • $begingroup$
                          @AccidentalFourierTransform: But here the answer is a meromorphic function of $p$, and we don't even have an accumulation point. So do you have a real proof? =)
                          $endgroup$
                          – user21820
                          Nov 25 '18 at 4:22










                        • $begingroup$
                          @user21820 Oh, no, not at all. I was just pointing out that Yidati's answer was only valid for integer $p$. I didn't want to suggest that I have a better answer.
                          $endgroup$
                          – AccidentalFourierTransform
                          Nov 25 '18 at 4:26
















                        $begingroup$
                        This only works for integral $p$, right?
                        $endgroup$
                        – AccidentalFourierTransform
                        Nov 24 '18 at 22:13




                        $begingroup$
                        This only works for integral $p$, right?
                        $endgroup$
                        – AccidentalFourierTransform
                        Nov 24 '18 at 22:13












                        $begingroup$
                        @AccidentalFourierTransform: Your comment makes me wonder whether there is a way to bootstrap from integer $p$ to the complex $p$...
                        $endgroup$
                        – user21820
                        Nov 25 '18 at 4:13




                        $begingroup$
                        @AccidentalFourierTransform: Your comment makes me wonder whether there is a way to bootstrap from integer $p$ to the complex $p$...
                        $endgroup$
                        – user21820
                        Nov 25 '18 at 4:13












                        $begingroup$
                        @user21820 Something in the spirit of the Identity theorem, or more generally Analytic continuation, I guess...
                        $endgroup$
                        – AccidentalFourierTransform
                        Nov 25 '18 at 4:16




                        $begingroup$
                        @user21820 Something in the spirit of the Identity theorem, or more generally Analytic continuation, I guess...
                        $endgroup$
                        – AccidentalFourierTransform
                        Nov 25 '18 at 4:16












                        $begingroup$
                        @AccidentalFourierTransform: But here the answer is a meromorphic function of $p$, and we don't even have an accumulation point. So do you have a real proof? =)
                        $endgroup$
                        – user21820
                        Nov 25 '18 at 4:22




                        $begingroup$
                        @AccidentalFourierTransform: But here the answer is a meromorphic function of $p$, and we don't even have an accumulation point. So do you have a real proof? =)
                        $endgroup$
                        – user21820
                        Nov 25 '18 at 4:22












                        $begingroup$
                        @user21820 Oh, no, not at all. I was just pointing out that Yidati's answer was only valid for integer $p$. I didn't want to suggest that I have a better answer.
                        $endgroup$
                        – AccidentalFourierTransform
                        Nov 25 '18 at 4:26




                        $begingroup$
                        @user21820 Oh, no, not at all. I was just pointing out that Yidati's answer was only valid for integer $p$. I didn't want to suggest that I have a better answer.
                        $endgroup$
                        – AccidentalFourierTransform
                        Nov 25 '18 at 4:26











                        0












                        $begingroup$

                        So let us follow your initial line of thought and convert the integral to a double integral. As you correctly observe, as
                        $$frac{1}{x^p} = frac{1}{Gamma (p)} int_0^infty e^{-xt} t^{p - 1} , dt,$$
                        which, by the way, is just the Laplace transform for the function $x^{p -1}$, as a double integral your integral can be rewritten as
                        $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty int_0^infty e^{-xt} cos x t^{p - 1} , dt , dx,$$
                        or
                        $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty t^{p - 1} int_0^infty e^{-xt} cos x , dx , dt,$$
                        after changing the order of integration.



                        The inner $x$-integral can be readily found. Either using integration by parts twice, or recognising the integral as the Laplace transform for the function $cos x$, as
                        $$int_0^infty e^{-xt} cos x , dx = frac{t}{1 + t^2},$$
                        we have
                        $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty frac{t^p}{1 + t^2} , dt.$$
                        Enforcing a substitution of $t mapsto sqrt{t}$ leads to
                        begin{align}
                        int_0^infty frac{cos x}{x^p} , dx &= frac{1}{2 Gamma (p)} int_0^infty frac{t^{frac{p}{2} - frac{1}{2}}}{1 + t} , dt = frac{1}{2 Gamma (p)} int_0^infty frac{t^{frac{p + 1}{2} - 1}}{(1 + t)^{frac{p + 1}{2} + frac{1 - p}{2}}}.
                        end{align}

                        As this is exactly of the form of the Beta function (see the second of the integral representations in the link) we have
                        begin{align}
                        int_0^infty frac{cos x}{x^p} , dx &= frac{1}{2 Gamma (p)} text{B} left (frac{p + 1}{2}, frac{1 - p}{2} right )\
                        &= frac{1}{2 Gamma (p)} Gamma left (frac{p}{2} + frac{1}{2} right ) Gamma left (frac{1}{2} - frac{p}{2} right ) \
                        &= frac{1}{2 Gamma (p)} Gamma left (frac{p}{2} + frac{1}{2} right ) Gamma left [1 - left (frac{p}{2} + frac{1}{2} right ) right ] \
                        &= frac{1}{2 Gamma (p)} frac{pi}{sin (p + 1)pi/2} qquad (*)\
                        &= frac{pi}{2 Gamma (p) cos left (frac{pi p}{2} right )},
                        end{align}

                        as required. Note Euler's reflection formula was used in ($*$).






                        share|cite|improve this answer









                        $endgroup$


















                          0












                          $begingroup$

                          So let us follow your initial line of thought and convert the integral to a double integral. As you correctly observe, as
                          $$frac{1}{x^p} = frac{1}{Gamma (p)} int_0^infty e^{-xt} t^{p - 1} , dt,$$
                          which, by the way, is just the Laplace transform for the function $x^{p -1}$, as a double integral your integral can be rewritten as
                          $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty int_0^infty e^{-xt} cos x t^{p - 1} , dt , dx,$$
                          or
                          $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty t^{p - 1} int_0^infty e^{-xt} cos x , dx , dt,$$
                          after changing the order of integration.



                          The inner $x$-integral can be readily found. Either using integration by parts twice, or recognising the integral as the Laplace transform for the function $cos x$, as
                          $$int_0^infty e^{-xt} cos x , dx = frac{t}{1 + t^2},$$
                          we have
                          $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty frac{t^p}{1 + t^2} , dt.$$
                          Enforcing a substitution of $t mapsto sqrt{t}$ leads to
                          begin{align}
                          int_0^infty frac{cos x}{x^p} , dx &= frac{1}{2 Gamma (p)} int_0^infty frac{t^{frac{p}{2} - frac{1}{2}}}{1 + t} , dt = frac{1}{2 Gamma (p)} int_0^infty frac{t^{frac{p + 1}{2} - 1}}{(1 + t)^{frac{p + 1}{2} + frac{1 - p}{2}}}.
                          end{align}

                          As this is exactly of the form of the Beta function (see the second of the integral representations in the link) we have
                          begin{align}
                          int_0^infty frac{cos x}{x^p} , dx &= frac{1}{2 Gamma (p)} text{B} left (frac{p + 1}{2}, frac{1 - p}{2} right )\
                          &= frac{1}{2 Gamma (p)} Gamma left (frac{p}{2} + frac{1}{2} right ) Gamma left (frac{1}{2} - frac{p}{2} right ) \
                          &= frac{1}{2 Gamma (p)} Gamma left (frac{p}{2} + frac{1}{2} right ) Gamma left [1 - left (frac{p}{2} + frac{1}{2} right ) right ] \
                          &= frac{1}{2 Gamma (p)} frac{pi}{sin (p + 1)pi/2} qquad (*)\
                          &= frac{pi}{2 Gamma (p) cos left (frac{pi p}{2} right )},
                          end{align}

                          as required. Note Euler's reflection formula was used in ($*$).






                          share|cite|improve this answer









                          $endgroup$
















                            0












                            0








                            0





                            $begingroup$

                            So let us follow your initial line of thought and convert the integral to a double integral. As you correctly observe, as
                            $$frac{1}{x^p} = frac{1}{Gamma (p)} int_0^infty e^{-xt} t^{p - 1} , dt,$$
                            which, by the way, is just the Laplace transform for the function $x^{p -1}$, as a double integral your integral can be rewritten as
                            $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty int_0^infty e^{-xt} cos x t^{p - 1} , dt , dx,$$
                            or
                            $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty t^{p - 1} int_0^infty e^{-xt} cos x , dx , dt,$$
                            after changing the order of integration.



                            The inner $x$-integral can be readily found. Either using integration by parts twice, or recognising the integral as the Laplace transform for the function $cos x$, as
                            $$int_0^infty e^{-xt} cos x , dx = frac{t}{1 + t^2},$$
                            we have
                            $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty frac{t^p}{1 + t^2} , dt.$$
                            Enforcing a substitution of $t mapsto sqrt{t}$ leads to
                            begin{align}
                            int_0^infty frac{cos x}{x^p} , dx &= frac{1}{2 Gamma (p)} int_0^infty frac{t^{frac{p}{2} - frac{1}{2}}}{1 + t} , dt = frac{1}{2 Gamma (p)} int_0^infty frac{t^{frac{p + 1}{2} - 1}}{(1 + t)^{frac{p + 1}{2} + frac{1 - p}{2}}}.
                            end{align}

                            As this is exactly of the form of the Beta function (see the second of the integral representations in the link) we have
                            begin{align}
                            int_0^infty frac{cos x}{x^p} , dx &= frac{1}{2 Gamma (p)} text{B} left (frac{p + 1}{2}, frac{1 - p}{2} right )\
                            &= frac{1}{2 Gamma (p)} Gamma left (frac{p}{2} + frac{1}{2} right ) Gamma left (frac{1}{2} - frac{p}{2} right ) \
                            &= frac{1}{2 Gamma (p)} Gamma left (frac{p}{2} + frac{1}{2} right ) Gamma left [1 - left (frac{p}{2} + frac{1}{2} right ) right ] \
                            &= frac{1}{2 Gamma (p)} frac{pi}{sin (p + 1)pi/2} qquad (*)\
                            &= frac{pi}{2 Gamma (p) cos left (frac{pi p}{2} right )},
                            end{align}

                            as required. Note Euler's reflection formula was used in ($*$).






                            share|cite|improve this answer









                            $endgroup$



                            So let us follow your initial line of thought and convert the integral to a double integral. As you correctly observe, as
                            $$frac{1}{x^p} = frac{1}{Gamma (p)} int_0^infty e^{-xt} t^{p - 1} , dt,$$
                            which, by the way, is just the Laplace transform for the function $x^{p -1}$, as a double integral your integral can be rewritten as
                            $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty int_0^infty e^{-xt} cos x t^{p - 1} , dt , dx,$$
                            or
                            $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty t^{p - 1} int_0^infty e^{-xt} cos x , dx , dt,$$
                            after changing the order of integration.



                            The inner $x$-integral can be readily found. Either using integration by parts twice, or recognising the integral as the Laplace transform for the function $cos x$, as
                            $$int_0^infty e^{-xt} cos x , dx = frac{t}{1 + t^2},$$
                            we have
                            $$int_0^infty frac{cos x}{x^p} , dx = frac{1}{Gamma (p)} int_0^infty frac{t^p}{1 + t^2} , dt.$$
                            Enforcing a substitution of $t mapsto sqrt{t}$ leads to
                            begin{align}
                            int_0^infty frac{cos x}{x^p} , dx &= frac{1}{2 Gamma (p)} int_0^infty frac{t^{frac{p}{2} - frac{1}{2}}}{1 + t} , dt = frac{1}{2 Gamma (p)} int_0^infty frac{t^{frac{p + 1}{2} - 1}}{(1 + t)^{frac{p + 1}{2} + frac{1 - p}{2}}}.
                            end{align}

                            As this is exactly of the form of the Beta function (see the second of the integral representations in the link) we have
                            begin{align}
                            int_0^infty frac{cos x}{x^p} , dx &= frac{1}{2 Gamma (p)} text{B} left (frac{p + 1}{2}, frac{1 - p}{2} right )\
                            &= frac{1}{2 Gamma (p)} Gamma left (frac{p}{2} + frac{1}{2} right ) Gamma left (frac{1}{2} - frac{p}{2} right ) \
                            &= frac{1}{2 Gamma (p)} Gamma left (frac{p}{2} + frac{1}{2} right ) Gamma left [1 - left (frac{p}{2} + frac{1}{2} right ) right ] \
                            &= frac{1}{2 Gamma (p)} frac{pi}{sin (p + 1)pi/2} qquad (*)\
                            &= frac{pi}{2 Gamma (p) cos left (frac{pi p}{2} right )},
                            end{align}

                            as required. Note Euler's reflection formula was used in ($*$).







                            share|cite|improve this answer












                            share|cite|improve this answer



                            share|cite|improve this answer










                            answered Jan 4 at 10:01









                            omegadotomegadot

                            5,5402728




                            5,5402728






























                                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%2f3012090%2fhow-to-evaluate-this-nonelementary-integral%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

                                Human spaceflight

                                Can not write log (Is /dev/pts mounted?) - openpty in Ubuntu-on-Windows?

                                張江高科駅