Double Integral: solution with hypergeometric function?

Click For Summary
SUMMARY

The discussion centers on evaluating the double integral $$\int_0^1 dv \int_0^1 dw \frac{(vw)^n(1-v)^m}{(1-vw)^\alpha}$$ with parameters where $$\Re(n),\Re(m) \geq 0$$ and $$\alpha = 1,2,3$$. The user references the Table of Integrals, Series and Products by Gradshteyn & Ryzhik, specifically equation (3.197.3), which relates the integral to the beta function and the ordinary hypergeometric function $$ _2F_1$$. The user expresses concern about the validity of this formula at the upper limit $$w=1$$, noting potential divergence issues, particularly when $$m=0$$, and seeks clarification on whether the entire integral may diverge.

PREREQUISITES
  • Understanding of double integrals and their properties.
  • Familiarity with the beta function and hypergeometric functions, specifically $$ _2F_1$$.
  • Knowledge of singularities in integrals and their implications on convergence.
  • Experience with reference materials like the Table of Integrals, Series and Products by Gradshteyn & Ryzhik.
NEXT STEPS
  • Research the properties of the beta function and its relationship to hypergeometric functions.
  • Study the conditions under which integrals diverge, particularly focusing on logarithmic singularities.
  • Explore techniques for evaluating integrals with singularities, including regularization methods.
  • Examine specific cases of double integrals involving hypergeometric functions for practical applications.
USEFUL FOR

Mathematicians, students studying advanced calculus, and researchers working with special functions and integral evaluations.

Dirickby
Messages
3
Reaction score
1

Homework Statement


Hello, I've recently encountered this double integral
$$\int_0^1 dv \int_0^1 dw \frac{(vw)^n(1-v)^m}{(1-vw)^\alpha} $$

with ## \Re(n),\Re(m) \geq 0## and ##\alpha = 1,2,3##.

Homework Equations



I use Table of Integrals, Series and Products by Gradshteyn & Ryzhik as a reference. There I found that
$$\int_0^1 dv \frac{(v)^n(1-v)^m}{(1-vw)^\alpha} = \beta(n+1, m+1)
_2F_1(\alpha ,n+1; n+m+2; w) \qquad\text{(3.197.3)} $$
with the ##\beta##-function and ##_2F_1## the ordinary hypergeometric function. But this equation is only valid for ##w<1## (and ## \Re(n+1),\Re(m+1) > 0##, which is the case here).

The Attempt at a Solution


If I could use the formula above, I could easily integrate over ##v## and then over ##w##, but I suppose it would be wrong as ##w=1## at the upper integration limit. Can I somehow bypass this?
Additionnally if we were to only consider ##w=1##, the integral would diverge at least in some cases (e.g. ##m=0##). Could my whole integral diverge because of this?
 
Physics news on Phys.org
At w=1, the integral from Gradstein diverges logarithmically. But the integral over a logarithmic singularity is finite, so the integral over w should exist.
 
  • Like
Likes   Reactions: Dirickby

Similar threads

Replies
2
Views
2K
Replies
3
Views
2K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
Replies
2
Views
2K
Replies
4
Views
2K
  • · Replies 5 ·
Replies
5
Views
1K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K