Poles of products of Gamma functions

+ 2 like - 0 dislike
126 views

I want to know if there can be a general statement about the poles (Laurent expansion) of such products of Gamma functions as a function of $p \in \mathbb{R}$ in the limit $\epsilon \rightarrow 0$,

$\Gamma[ \frac{(-1-2p)}{2} - \epsilon ]\Gamma[\frac{-1+p}{2} + \epsilon]\Gamma[\frac{(5+p)}{2} + \epsilon]$

For any given $p$ such that the $p$ dependent part of the argument is a negative integer one can do the usual Laurent expansion of the Gamma function in $\epsilon$ for each of the 3 factors and then multiply. But what can be said about the Laurent expansion in general as a function of $p$?

I wish one could write down the Laurent expansion in $\epsilon$ as a function of $p$!

One sees that there are these special cases like if $p$ is such that there are two integers $N$ and $M$ satisfying, $p= 1-2N^2 = -5 -2M^2$ then the later two Gamma function can have poles simultaneously. (like $N=2, M =1, p= -7$) Existence of such special $p$ naively seems to make things more tricky.

This post imported from StackExchange MathOverflow at 2014-08-29 08:16 (UCT), posted by SE-user Anirbit
retagged Aug 29, 2014
Are you missing a square root in the definition of your product? Otherwise I can't imagine why it should matter that the argument of $\Gamma$ is near a square.

This post imported from StackExchange MathOverflow at 2014-08-29 08:16 (UCT), posted by SE-user Greg Martin
@Greg Martin Are you referring to the fact that I have written $N^2$ and $M^2$? That I did so that, $(p-1)/2 = - N^2$ is a negative definite integer and hence the second Gamma function has a pole. Similarly, $(5+p)/2 = -M^2$ is again a negative definite integer and hence the 3rd Gamma function has a pole. It would be great if you can help with the question.

This post imported from StackExchange MathOverflow at 2014-08-29 08:16 (UCT), posted by SE-user Anirbit
Anirbit: the Gamma function has poles for every non-positive integer argument. Squares do not seem to enter the picture anywhere.

This post imported from StackExchange MathOverflow at 2014-08-29 08:16 (UCT), posted by SE-user S. Carnahan
@S.Carnahan As you can see in my comment above to Greg, I have explained why I parametrized using squares - to ensure negative definiteness.

This post imported from StackExchange MathOverflow at 2014-08-29 08:16 (UCT), posted by SE-user Anirbit

Forget the $p$'s and consider a product $\Gamma(x-\epsilon)\Gamma(y-\epsilon)\ldots\Gamma(z-\epsilon)$. If one of $x,y,\ldots,z$ is not a negative integer, it is regular, and you can put $\epsilon=0$ in the factor. So it is enough to consider the case where all of $x,y,\ldots,z$ are negative integers. In this case, if there are $k$ factors,  $\epsilon^k\Gamma(x-\epsilon)\Gamma(y-\epsilon)\ldots\Gamma(z-\epsilon)$ is analytic at $\epsilon=0$.
Therefore you can expand it into a power series using the power series for  $\epsilon\Gamma(x-\epsilon)$ etc. to order $k$ or higher. At the end substitute for $x,y,\ldots,z$ your expressions in $p$,
 Please use answers only to (at least partly) answer questions. To comment, discuss, or ask for clarification, leave a comment instead. To mask links under text, please type your text, highlight it, and click the "link" button. You can then enter your link URL. Please consult the FAQ for as to how to format your post. This is the answer box; if you want to write a comment instead, please use the 'add comment' button. Live preview (may slow down editor)   Preview Your name to display (optional): Email me at this address if my answer is selected or commented on: Privacy: Your email address will only be used for sending these notifications. Anti-spam verification: If you are a human please identify the position of the character covered by the symbol $\varnothing$ in the following word:p$\hbar$ysicsOver$\varnothing$lowThen drag the red bullet below over the corresponding character of our banner. When you drop it there, the bullet changes to green (on slow internet connections after a few seconds). To avoid this verification in future, please log in or register.