Graduate Is this a known result? (hypergeometric function at special values)

  • Thread starter Thread starter nrqed
  • Start date Start date
  • Tags Tags
    Function
Click For Summary
SUMMARY

The forum discussion centers on the verification of the hypergeometric function identity $$ \,_2F_1 [\frac{1}{2},-N;-N+ \frac{1}{2} ; z=1] = \frac{4^N (N!)^2}{(2N)!} $$ for positive integers N. Users reference the Digital Library of Mathematical Functions (DLMF) and specific identities such as 15.4.24 and Pochhammer symbols to derive the result. They also discuss the need for additional resources on generalized hypergeometric functions, particularly for identities involving parameters beyond the standard forms.

PREREQUISITES
  • Understanding of hypergeometric functions, specifically the notation and properties of $$ _pF_q $$.
  • Familiarity with Pochhammer symbols and their definitions, particularly $$ (a)_N = \Gamma(a+N)/\Gamma(a) $$.
  • Knowledge of the Digital Library of Mathematical Functions (DLMF) and its resources.
  • Basic proficiency in mathematical functions and identities, including those found in Erdelyi's "Higher Transcendental Functions".
NEXT STEPS
  • Research the specific identities listed in DLMF sections 5.2.5, 5.2.6, 5.2.7, 5.4.6, and 5.5.5.
  • Explore Wolfram Mathworld for additional relations involving generalized hypergeometric functions.
  • Investigate Erdelyi's "Higher Transcendental Functions" for further identities and their applications.
  • Examine Yudell Luke's works on special functions for advanced insights into mathematical approximations.
USEFUL FOR

Mathematicians, researchers in mathematical functions, and students studying hypergeometric functions and their applications will benefit from this discussion.

nrqed
Science Advisor
Messages
3,762
Reaction score
297
Through some calculations in a graph counting problem, I have checked that for many values of N (a positive integer), the following is true:

$$ \,_2F_1 [\frac{1}{2},-N;-N+ \frac{1}{2} ; z=1] = \frac{4^N (N!)^2}{(2N)!} $$
I would like to prove that this correct for arbitrary N, but I cannot find anywhere an expression for this particular combination of parameters (note that ##c-a-b=0##, which is very unfortunate as it prevents the use of one well known identity). I have checked Abramowitz and Stegun and several other sources. No luck. Does anyone know if this is true or if they have some good source of identities to recommend?
 
Physics news on Phys.org
It is a result that follows very easily using identities available at the DLMF:
https://dlmf.nist.gov/

The starting point is 15.4.24 (http://dlmf.nist.gov/15.4.E24)
## _2F_1\left(-N,b;c;1\right)=\frac{{\left(c-b\right)_{N}}}{{\left(c\right)_{N}}} ##
where the Pochhammer symbols are ##(a)_N = \Gamma(a+N)/\Gamma(a)##. You then use a few identities for those symbols to get your result. I used 5.2.7, 5.2.6, 5.2.5, 5.5.5 and 5.4.6 (see https://dlmf.nist.gov/5) .

jason
 
  • Like
Likes jim mcnamara, nrqed and DrClaude
jasonRF said:
It is a result that follows very easily using identities available at the DLMF:
https://dlmf.nist.gov/

The starting point is 15.4.24 (http://dlmf.nist.gov/15.4.E24)
## _2F_1\left(-N,b;c;1\right)=\frac{{\left(c-b\right)_{N}}}{{\left(c\right)_{N}}} ##
where the Pochhammer symbols are ##(a)_N = \Gamma(a+N)/\Gamma(a)##. You then use a few identities for those symbols to get your result. I used 5.2.7, 5.2.6, 5.2.5, 5.5.5 and 5.4.6 (see https://dlmf.nist.gov/5) .

jason
Wow,

Thank you so much! I had seen the equation given in 15.4.24 but I had not realized that one could define ##(-n)_n## as is given in 5.2.7.

Thanks!
 
Last edited:
jasonRF said:
It is a result that follows very easily using identities available at the DLMF:
https://dlmf.nist.gov/

The starting point is 15.4.24 (http://dlmf.nist.gov/15.4.E24)
## _2F_1\left(-N,b;c;1\right)=\frac{{\left(c-b\right)_{N}}}{{\left(c\right)_{N}}} ##
where the Pochhammer symbols are ##(a)_N = \Gamma(a+N)/\Gamma(a)##. You then use a few identities for those symbols to get your result. I used 5.2.7, 5.2.6, 5.2.5, 5.5.5 and 5.4.6 (see https://dlmf.nist.gov/5) .

jason
Hello Jason,

I am now obtaining many conjectured identities that I cannot find in the literature. These involve generalized hypergeometric functions ##\,_pF_q## with ## p > 2## and ## q > 1##. For example I get a relation between ##\,_4F_3## and ##\,_3 F_2 ## for a certain choice of parameters (my relation involves three different positive integers).

Do you know if there are tables of known relations between generalized hypergeometric functions? I have seen a couple on the Wikipedia page but that's all.

Thank you!
 
Besides DLMF, I would check Wolfram Mathworld online. In terms of books, Erdelyi's "Higher Transcendental Functions" volumes 1-3 are legally available online:
http://numerical.recipes/oldverswitcher.html
They probably have a few relations.

I suspect that you don't have access to a library these days, but if you do then there may be other books to look at. For example, Yudell Luke's books "The special functions and their approximations" vol 1-2, and "mathematical functions and their approximations" should have some interesting information.

jason
 
  • Like
Likes nrqed
jasonRF said:
Besides DLMF, I would check Wolfram Mathworld online. In terms of books, Erdelyi's "Higher Transcendental Functions" volumes 1-3 are legally available online:
http://numerical.recipes/oldverswitcher.html
They probably have a few relations.

I suspect that you don't have access to a library these days, but if you do then there may be other books to look at. For example, Yudell Luke's books "The special functions and their approximations" vol 1-2, and "mathematical functions and their approximations" should have some interesting information.

jason
Thank you again, Jason. I appreciate it.
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
902
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 36 ·
2
Replies
36
Views
5K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K