Loosely related to my last question, I was trying to find a closed form of the finite sum
an:=n∑k=0(2kk)(−14)k
This is not too different from the well-known expression
n∑k=0(2kk)(14)k=(n+12n)=2n+122n(2nn)
so
n∑k=0(2kk)(−14)k=n∑k=0(2kk)(14)k−2n∑k=0k odd(2kk)(−14)k=2n+122n(2nn)−12⌊n−12⌋∑l=0(4l+22l+1)116l.
However, the second sum does not seem to be easier to handle at first glance. Also trying a similar ansatz an=(2kk)pn/22n led to seemingly unstructured pn given by (starting from p=0):
1,1,73,95,7235,97,18577,227143,57772145,…
So I was wondering (a) if there already is a known closed form for the an (I've searched quite a bit but sadly wasn't successful) and if not, then (b) what is a good way to tackle this problem - or does a "simple" (sum-free) form maybe not even exist?
Thanks in advance for any answer or comment!
Edit: I am aware of the solution
an=(−1)n2−2n+2(2n+2n+1)2F1(1;n+32;n+2;−1)+1√2
Mathematica presents where 2F1 is the hypergeometric function. But as the latter is given by an infinite sum, I was hoping for something simpler by connecting it to the known, non-alternating problem as described above - although I'd also accept if this was not possible.
Answer
(−1)k4k(2kk)=[xk]1√1+x
implies that
n∑k=0(−1)k4k(2kk)=[xn]1(1−x)√1+x=1√2+1√2[xn]11+x+√2+2x
hence an is positive and convergent to 1√2, and the hypergeometric representation is straightforward.
The Maclaurin coefficients of g(x)=1(1−x)+√2√1−x are positive and they behave like 1√2πn for large values of n, hence we have the approximate identity
an≈1√2+(−1)n2√πn
for n→+∞.
No comments:
Post a Comment