Closed expressions for divergent series over Bernoulli numbers?

The name of the pictureThe name of the pictureThe name of the pictureClash Royale CLAN TAG#URR8PPP











up vote
3
down vote

favorite
1












Motivation



In a recent post (Asymptotic behaviour of sums involving $k$, $log(k)$ and $H_k$) I asked for the asymptotic behaviour of the sum



$$sigma_c(n)=sum_k=1^n H_k log(k)tag1$$



and I found that the constant in the asymptotic expression contained, among other known constants, the sum



$$kappa_c = sum_m=1^infty fracB(2m)2m zeta'(2m)\=frac112 zeta '(2)-fraczeta '(4)120+fraczeta '(6)252-+...tag2$$



where $B(n)$ is the Bernoulli number of order $n$ and $zeta'()$ is the derivative of the Riemann zeta function. This is a strongly divergent series, and I had to resort to the limit



$$lim_ntoinfty (sigma_c(n) - (textleading terms))tag3$$



for calculating the complete constant, a sum of known constants and $kappa_c$, which, however, I could only use numerically.



Quite recently, however, in an answer to Constant term in Stirling type formula for $sum^N_n=1 H_n cdot ln(n)$, a consistent interpretation of the sum in (2) was given in terms of a convergent integral and the numerical value calculated.



Questions



This motivated me to ask for the values of similar divergent series, and for a proof of the interpretation of the following two examples:



$$kappa_d :=sum_m=1^infty B(2m)dot= fracpi^26-frac32simeq 0.144934$$



and



$$kappa_e :=sum_m=1^infty fracB(2m)2mdot=gamma -frac12simeq 0.0772157 $$



Here $gamma$ is the Euler-Mascheroni constant and $dot=$ means "is understood as" in the Ramanujan sense that $1+2+3+... dot= -frac112$










share|cite|improve this question



























    up vote
    3
    down vote

    favorite
    1












    Motivation



    In a recent post (Asymptotic behaviour of sums involving $k$, $log(k)$ and $H_k$) I asked for the asymptotic behaviour of the sum



    $$sigma_c(n)=sum_k=1^n H_k log(k)tag1$$



    and I found that the constant in the asymptotic expression contained, among other known constants, the sum



    $$kappa_c = sum_m=1^infty fracB(2m)2m zeta'(2m)\=frac112 zeta '(2)-fraczeta '(4)120+fraczeta '(6)252-+...tag2$$



    where $B(n)$ is the Bernoulli number of order $n$ and $zeta'()$ is the derivative of the Riemann zeta function. This is a strongly divergent series, and I had to resort to the limit



    $$lim_ntoinfty (sigma_c(n) - (textleading terms))tag3$$



    for calculating the complete constant, a sum of known constants and $kappa_c$, which, however, I could only use numerically.



    Quite recently, however, in an answer to Constant term in Stirling type formula for $sum^N_n=1 H_n cdot ln(n)$, a consistent interpretation of the sum in (2) was given in terms of a convergent integral and the numerical value calculated.



    Questions



    This motivated me to ask for the values of similar divergent series, and for a proof of the interpretation of the following two examples:



    $$kappa_d :=sum_m=1^infty B(2m)dot= fracpi^26-frac32simeq 0.144934$$



    and



    $$kappa_e :=sum_m=1^infty fracB(2m)2mdot=gamma -frac12simeq 0.0772157 $$



    Here $gamma$ is the Euler-Mascheroni constant and $dot=$ means "is understood as" in the Ramanujan sense that $1+2+3+... dot= -frac112$










    share|cite|improve this question

























      up vote
      3
      down vote

      favorite
      1









      up vote
      3
      down vote

      favorite
      1






      1





      Motivation



      In a recent post (Asymptotic behaviour of sums involving $k$, $log(k)$ and $H_k$) I asked for the asymptotic behaviour of the sum



      $$sigma_c(n)=sum_k=1^n H_k log(k)tag1$$



      and I found that the constant in the asymptotic expression contained, among other known constants, the sum



      $$kappa_c = sum_m=1^infty fracB(2m)2m zeta'(2m)\=frac112 zeta '(2)-fraczeta '(4)120+fraczeta '(6)252-+...tag2$$



      where $B(n)$ is the Bernoulli number of order $n$ and $zeta'()$ is the derivative of the Riemann zeta function. This is a strongly divergent series, and I had to resort to the limit



      $$lim_ntoinfty (sigma_c(n) - (textleading terms))tag3$$



      for calculating the complete constant, a sum of known constants and $kappa_c$, which, however, I could only use numerically.



      Quite recently, however, in an answer to Constant term in Stirling type formula for $sum^N_n=1 H_n cdot ln(n)$, a consistent interpretation of the sum in (2) was given in terms of a convergent integral and the numerical value calculated.



      Questions



      This motivated me to ask for the values of similar divergent series, and for a proof of the interpretation of the following two examples:



      $$kappa_d :=sum_m=1^infty B(2m)dot= fracpi^26-frac32simeq 0.144934$$



      and



      $$kappa_e :=sum_m=1^infty fracB(2m)2mdot=gamma -frac12simeq 0.0772157 $$



      Here $gamma$ is the Euler-Mascheroni constant and $dot=$ means "is understood as" in the Ramanujan sense that $1+2+3+... dot= -frac112$










      share|cite|improve this question















      Motivation



      In a recent post (Asymptotic behaviour of sums involving $k$, $log(k)$ and $H_k$) I asked for the asymptotic behaviour of the sum



      $$sigma_c(n)=sum_k=1^n H_k log(k)tag1$$



      and I found that the constant in the asymptotic expression contained, among other known constants, the sum



      $$kappa_c = sum_m=1^infty fracB(2m)2m zeta'(2m)\=frac112 zeta '(2)-fraczeta '(4)120+fraczeta '(6)252-+...tag2$$



      where $B(n)$ is the Bernoulli number of order $n$ and $zeta'()$ is the derivative of the Riemann zeta function. This is a strongly divergent series, and I had to resort to the limit



      $$lim_ntoinfty (sigma_c(n) - (textleading terms))tag3$$



      for calculating the complete constant, a sum of known constants and $kappa_c$, which, however, I could only use numerically.



      Quite recently, however, in an answer to Constant term in Stirling type formula for $sum^N_n=1 H_n cdot ln(n)$, a consistent interpretation of the sum in (2) was given in terms of a convergent integral and the numerical value calculated.



      Questions



      This motivated me to ask for the values of similar divergent series, and for a proof of the interpretation of the following two examples:



      $$kappa_d :=sum_m=1^infty B(2m)dot= fracpi^26-frac32simeq 0.144934$$



      and



      $$kappa_e :=sum_m=1^infty fracB(2m)2mdot=gamma -frac12simeq 0.0772157 $$



      Here $gamma$ is the Euler-Mascheroni constant and $dot=$ means "is understood as" in the Ramanujan sense that $1+2+3+... dot= -frac112$







      sequences-and-series divergent-series bernoulli-numbers






      share|cite|improve this question















      share|cite|improve this question













      share|cite|improve this question




      share|cite|improve this question








      edited Aug 31 at 9:29

























      asked Aug 31 at 7:48









      Dr. Wolfgang Hintze

      2,575515




      2,575515




















          1 Answer
          1






          active

          oldest

          votes

















          up vote
          2
          down vote













          In both, use the following formula, available as Gradshteyn&Rhyzik 3.411.2
          $$fracB_2n2n = (-1)^n+1,2 int_0^infty fracx^2n-1e^2pi,x-1,dx $$
          Then a convergent integral for the second sum is derived by
          $$ sum_n=1^infty fracB_2n2n ,dot=
          2 int_0^infty fracdx/xe^2pi,x-1,sum_n=1^infty (-1)^n+1x^2n
          =2 int_0^infty fracdx,x(x^2+1)(e^2pi,x-1).$$
          Mathematica knows this integral, although G&R 3.415.1 can also be used:
          $$ (A) quad int_0^infty fracdx,x(x^2+b^2)(e^2pi,x-1) =
          frac12big( logb - psi(b) - frac12b big) $$
          For $b=1$ we therefore have
          $$ sum_n=1^infty fracB_2n2n ,dot= ,gamma - frac12 .$$
          The other sum is similar,
          $$ sum_n=1^infty B_2n ,dot=
          4 int_0^infty fracdx/xe^2pi,x-1,sum_n=1^infty (-1)^n+1n,x^2n
          =4 int_0^infty fracdx,x(x^2+1)^2(e^2pi,x-1).$$
          Differentiate eq. $(A)$ and set $b=1$ to find
          $$ sum_n=1^infty B_2n,dot= ,fracpi^26 - frac32 .$$






          share|cite|improve this answer




















          • @ skbmoore Very good. I came from the "opposite direction" playing with Euler-Maclarin to solve the Basel problem and to calculate the asymptotics of $H_n$. (This was a nice exercise and I will present it as a self solution later; now I'm waiting for possible other solutions).
            – Dr. Wolfgang Hintze
            Sep 2 at 13:59











          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',
          convertImagesToLinks: true,
          noModals: false,
          showLowRepImageUploadWarning: true,
          reputationToPostImages: 10,
          bindNavPrevention: true,
          postfix: "",
          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%2f2900424%2fclosed-expressions-for-divergent-series-over-bernoulli-numbers%23new-answer', 'question_page');

          );

          Post as a guest






























          1 Answer
          1






          active

          oldest

          votes








          1 Answer
          1






          active

          oldest

          votes









          active

          oldest

          votes






          active

          oldest

          votes








          up vote
          2
          down vote













          In both, use the following formula, available as Gradshteyn&Rhyzik 3.411.2
          $$fracB_2n2n = (-1)^n+1,2 int_0^infty fracx^2n-1e^2pi,x-1,dx $$
          Then a convergent integral for the second sum is derived by
          $$ sum_n=1^infty fracB_2n2n ,dot=
          2 int_0^infty fracdx/xe^2pi,x-1,sum_n=1^infty (-1)^n+1x^2n
          =2 int_0^infty fracdx,x(x^2+1)(e^2pi,x-1).$$
          Mathematica knows this integral, although G&R 3.415.1 can also be used:
          $$ (A) quad int_0^infty fracdx,x(x^2+b^2)(e^2pi,x-1) =
          frac12big( logb - psi(b) - frac12b big) $$
          For $b=1$ we therefore have
          $$ sum_n=1^infty fracB_2n2n ,dot= ,gamma - frac12 .$$
          The other sum is similar,
          $$ sum_n=1^infty B_2n ,dot=
          4 int_0^infty fracdx/xe^2pi,x-1,sum_n=1^infty (-1)^n+1n,x^2n
          =4 int_0^infty fracdx,x(x^2+1)^2(e^2pi,x-1).$$
          Differentiate eq. $(A)$ and set $b=1$ to find
          $$ sum_n=1^infty B_2n,dot= ,fracpi^26 - frac32 .$$






          share|cite|improve this answer




















          • @ skbmoore Very good. I came from the "opposite direction" playing with Euler-Maclarin to solve the Basel problem and to calculate the asymptotics of $H_n$. (This was a nice exercise and I will present it as a self solution later; now I'm waiting for possible other solutions).
            – Dr. Wolfgang Hintze
            Sep 2 at 13:59















          up vote
          2
          down vote













          In both, use the following formula, available as Gradshteyn&Rhyzik 3.411.2
          $$fracB_2n2n = (-1)^n+1,2 int_0^infty fracx^2n-1e^2pi,x-1,dx $$
          Then a convergent integral for the second sum is derived by
          $$ sum_n=1^infty fracB_2n2n ,dot=
          2 int_0^infty fracdx/xe^2pi,x-1,sum_n=1^infty (-1)^n+1x^2n
          =2 int_0^infty fracdx,x(x^2+1)(e^2pi,x-1).$$
          Mathematica knows this integral, although G&R 3.415.1 can also be used:
          $$ (A) quad int_0^infty fracdx,x(x^2+b^2)(e^2pi,x-1) =
          frac12big( logb - psi(b) - frac12b big) $$
          For $b=1$ we therefore have
          $$ sum_n=1^infty fracB_2n2n ,dot= ,gamma - frac12 .$$
          The other sum is similar,
          $$ sum_n=1^infty B_2n ,dot=
          4 int_0^infty fracdx/xe^2pi,x-1,sum_n=1^infty (-1)^n+1n,x^2n
          =4 int_0^infty fracdx,x(x^2+1)^2(e^2pi,x-1).$$
          Differentiate eq. $(A)$ and set $b=1$ to find
          $$ sum_n=1^infty B_2n,dot= ,fracpi^26 - frac32 .$$






          share|cite|improve this answer




















          • @ skbmoore Very good. I came from the "opposite direction" playing with Euler-Maclarin to solve the Basel problem and to calculate the asymptotics of $H_n$. (This was a nice exercise and I will present it as a self solution later; now I'm waiting for possible other solutions).
            – Dr. Wolfgang Hintze
            Sep 2 at 13:59













          up vote
          2
          down vote










          up vote
          2
          down vote









          In both, use the following formula, available as Gradshteyn&Rhyzik 3.411.2
          $$fracB_2n2n = (-1)^n+1,2 int_0^infty fracx^2n-1e^2pi,x-1,dx $$
          Then a convergent integral for the second sum is derived by
          $$ sum_n=1^infty fracB_2n2n ,dot=
          2 int_0^infty fracdx/xe^2pi,x-1,sum_n=1^infty (-1)^n+1x^2n
          =2 int_0^infty fracdx,x(x^2+1)(e^2pi,x-1).$$
          Mathematica knows this integral, although G&R 3.415.1 can also be used:
          $$ (A) quad int_0^infty fracdx,x(x^2+b^2)(e^2pi,x-1) =
          frac12big( logb - psi(b) - frac12b big) $$
          For $b=1$ we therefore have
          $$ sum_n=1^infty fracB_2n2n ,dot= ,gamma - frac12 .$$
          The other sum is similar,
          $$ sum_n=1^infty B_2n ,dot=
          4 int_0^infty fracdx/xe^2pi,x-1,sum_n=1^infty (-1)^n+1n,x^2n
          =4 int_0^infty fracdx,x(x^2+1)^2(e^2pi,x-1).$$
          Differentiate eq. $(A)$ and set $b=1$ to find
          $$ sum_n=1^infty B_2n,dot= ,fracpi^26 - frac32 .$$






          share|cite|improve this answer












          In both, use the following formula, available as Gradshteyn&Rhyzik 3.411.2
          $$fracB_2n2n = (-1)^n+1,2 int_0^infty fracx^2n-1e^2pi,x-1,dx $$
          Then a convergent integral for the second sum is derived by
          $$ sum_n=1^infty fracB_2n2n ,dot=
          2 int_0^infty fracdx/xe^2pi,x-1,sum_n=1^infty (-1)^n+1x^2n
          =2 int_0^infty fracdx,x(x^2+1)(e^2pi,x-1).$$
          Mathematica knows this integral, although G&R 3.415.1 can also be used:
          $$ (A) quad int_0^infty fracdx,x(x^2+b^2)(e^2pi,x-1) =
          frac12big( logb - psi(b) - frac12b big) $$
          For $b=1$ we therefore have
          $$ sum_n=1^infty fracB_2n2n ,dot= ,gamma - frac12 .$$
          The other sum is similar,
          $$ sum_n=1^infty B_2n ,dot=
          4 int_0^infty fracdx/xe^2pi,x-1,sum_n=1^infty (-1)^n+1n,x^2n
          =4 int_0^infty fracdx,x(x^2+1)^2(e^2pi,x-1).$$
          Differentiate eq. $(A)$ and set $b=1$ to find
          $$ sum_n=1^infty B_2n,dot= ,fracpi^26 - frac32 .$$







          share|cite|improve this answer












          share|cite|improve this answer



          share|cite|improve this answer










          answered Sep 2 at 4:03









          skbmoore

          1,49929




          1,49929











          • @ skbmoore Very good. I came from the "opposite direction" playing with Euler-Maclarin to solve the Basel problem and to calculate the asymptotics of $H_n$. (This was a nice exercise and I will present it as a self solution later; now I'm waiting for possible other solutions).
            – Dr. Wolfgang Hintze
            Sep 2 at 13:59

















          • @ skbmoore Very good. I came from the "opposite direction" playing with Euler-Maclarin to solve the Basel problem and to calculate the asymptotics of $H_n$. (This was a nice exercise and I will present it as a self solution later; now I'm waiting for possible other solutions).
            – Dr. Wolfgang Hintze
            Sep 2 at 13:59
















          @ skbmoore Very good. I came from the "opposite direction" playing with Euler-Maclarin to solve the Basel problem and to calculate the asymptotics of $H_n$. (This was a nice exercise and I will present it as a self solution later; now I'm waiting for possible other solutions).
          – Dr. Wolfgang Hintze
          Sep 2 at 13:59





          @ skbmoore Very good. I came from the "opposite direction" playing with Euler-Maclarin to solve the Basel problem and to calculate the asymptotics of $H_n$. (This was a nice exercise and I will present it as a self solution later; now I'm waiting for possible other solutions).
          – Dr. Wolfgang Hintze
          Sep 2 at 13:59


















           

          draft saved


          draft discarded















































           


          draft saved


          draft discarded














          StackExchange.ready(
          function ()
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f2900424%2fclosed-expressions-for-divergent-series-over-bernoulli-numbers%23new-answer', 'question_page');

          );

          Post as a guest













































































          這個網誌中的熱門文章

          How to combine Bézier curves to a surface?

          Mutual Information Always Non-negative

          Why am i infinitely getting the same tweet with the Twitter Search API?