$$ 1-5\left(\frac{1}{2}\right)^3+9\left(\frac{1 \cdot 3}{2 \cdot 4}\right)^3-13\left(\frac{1 \cdot 3 \cdot 5}{2 \cdot 4 \cdot 6}\right)^3+\cdots $$

I went on evaluating the above series and encountered that solving $\displaystyle \sum_{n\ge 0}\left(\binom{2n}{n}\right)^3x^n$ would suffice.

But how do we make a generating function for the third power of a central binomial coefficient using the fact $\displaystyle \sum_{n\ge 0}\binom{2n}{n}x^n=\frac{1}{\sqrt{1-4x}}$

  • $\begingroup$ The generating function is not algebraic. It can be expressed using a hypergeometric function: see OEIS sequence A002897. But that hypergeometric function is unlikely to help you. $\endgroup$ – Robert Israel Jan 13 '17 at 5:23
  • $\begingroup$ The "power 2" generating function is in terms of elliptic integral and this one as you say in terms of hypergeometric functions, so there isn't any closed form I guess. But Ramanujan did find a value to the series I posted above. It was like $\frac{2}{\pi}$ as per my memory. $\endgroup$ – Aditya Narayan Sharma Jan 13 '17 at 5:27

This is not an answer but just a result obtained using a CAS.

Let $$f_k=\sum_{n=0}^\infty \binom{2 n}{n}^kx^n$$ The following expressions have been obtained $$f_1=\frac{1}{\sqrt{1-4 x}}$$ $$f_2=\frac{2 }{\pi }K(16 x)$$ $$f_3=\frac{4 }{\pi ^2}K\left(\frac{1}{2} \left(1-\sqrt{1-64 x}\right)\right)^2$$ $$f_4=\, _4F_3\left(\frac{1}{2},\frac{1}{2},\frac{1}{2},\frac{1}{2};1,1,1;256 x\right)$$ $$f_5=\, _5F_4\left(\frac{1}{2},\frac{1}{2},\frac{1}{2},\frac{1}{2},\frac{1}{2};1,1,1,1;1 024 x\right)$$ where appear, for $k=2,3$, the complete elliptic integrals of the first kind and, for $k>3$, the generalized hypergeometric functions

  • $\begingroup$ "+1" , Thanks for the additional info , Is there any proof of these 5 over the web ? $\endgroup$ – Aditya Narayan Sharma Jan 13 '17 at 5:48
  • $\begingroup$ @AdityaNarayanSharma. I did not search for that. After your comment, just out of curiosity, I gave a CAs to eat ! Cheers. $\endgroup$ – Claude Leibovici Jan 13 '17 at 5:50
  • $\begingroup$ In the third case , the argument inside K is squared right ? And not the entire K-function ? $\endgroup$ – Aditya Narayan Sharma Jan 13 '17 at 5:53
  • $\begingroup$ @AdityaNarayanSharma. It is the entire function which is squared. $\endgroup$ – Claude Leibovici Jan 13 '17 at 8:10
  • 3
    $\begingroup$ Rewriting both my comment and your answer in terms of $~x=\bigg[\dfrac{\sin(2a)}{2^k}\bigg]^2,~$ with $~|a|<\dfrac\pi4,~$ we have $$F_0~=~\sec^2(2a),\quad F_1~=~\sec(2a),\quad F_2~=~\dfrac2\pi~K\Big(\sin(2a)\Big),\quad F_3~=~\bigg[\dfrac2\pi~K(\sin a)\bigg]^2.$$ Note: I've used the traditional modulus notation $K(k)$ for the argument of elliptic integrals, instead of the Mathematica one in terms the parameter $m=k^2.~$ $\endgroup$ – Lucian Jan 13 '17 at 9:33

This is one of the simplest and famous series given by Ramanujan and it's value is $2/\pi$. Unfortunately Ramanujan's technique requires a reasonable amount of effort to understand. I have presented the proof for this series and it's friend $$\frac{4}{\pi}= 1+\frac{7}{4}\left(\frac{1}{2}\right)^{3}+\frac{13}{4^{2}}\left(\frac{1\cdot 3}{2\cdot 4}\right)^{3}+\dots$$ in this post.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.