Sterling approximation of Beta Function

Click For Summary

Homework Help Overview

The discussion revolves around the application of Stirling's approximation to the Beta function, particularly focusing on the limit behavior of the expression involving the Gamma function as \( n \) approaches infinity. Participants are exploring how the terms behave under different conditions of \( x \).

Discussion Character

  • Exploratory, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants discuss the complexity of the problem and the challenges posed by the \( n^x \) term as \( n \) approaches infinity. There are attempts to rearrange the expression and apply Stirling's approximation to simplify the Gamma functions involved. Questions arise regarding the impact of \( x \) being a positive or negative integer or zero on the outcome.

Discussion Status

Several participants have provided insights and suggestions for simplifying the problem, including the use of a modified Stirling formula. There is acknowledgment of the importance of recognizing specific limits and behaviors of the functions involved, though no consensus has been reached on a final approach.

Contextual Notes

Participants are navigating the complexities of the problem while adhering to homework constraints, which may limit the information available for a complete resolution. The discussion reflects a variety of interpretations and methods being explored without definitive conclusions.

dykuma
Messages
55
Reaction score
6

Homework Statement


upload_2016-11-1_15-37-21.png

Homework Equations


upload_2016-11-1_15-37-31.png

upload_2016-11-1_15-37-41.png

The Attempt at a Solution


I think this problem is probably a lot simpler than I am making it out to be. However, when I apply sterling's approx., I get a very nasty equation that does not simplify easily.
upload_2016-11-1_15-39-48.png


One of the biggest problems I have though is dealing with that n^x term. it seems to me that as n approaches infinity, the outcome depends heavily on whether x is a positive or negative integer, or if its zero.
 
Physics news on Phys.org
Rearrange it to:

n^x B(x,n) =\Gamma(x) n^x \frac{\Gamma(n)}{\Gamma(n+x)}

So what you're really interested in is the limit of

n^x \frac{\Gamma(n)}{\Gamma(n+x)}

Use Sterling's approximation:

\Gamma(n) \approx n^{n+\frac{1}{2}} \sqrt{2 \pi} e^{-n}
\Gamma(n+x) \approx (n+x)^{n+x+\frac{1}{2}} \sqrt{2 \pi} e^{-(n+x)}

Write (n+x)^{n+x+\frac{1}{2}} as (n(1+\frac{x}{n}))^{n+x +\frac{1}{2}} = n^{n+x+\frac{1}{2}} (1+\frac{x}{n})^{n+x+\frac{1}{2}}
= n^{n+\frac{1}{2}} n^x (1+\frac{x}{n})^{n} (1+\frac{x}{n})^{x+\frac{1}{2}}

So
\Gamma(n+x) \approx n^{n+\frac{1}{2}} n^x (1+\frac{x}{n})^{n} (1+\frac{x}{n})^{x+\frac{1}{2}} \sqrt{2 \pi} e^{-(n+x)}

Now, you can just use a fact about e: e^x = lim_{n \rightarrow \infty} (1+\frac{x}{n})^n
 
  • Like
Likes   Reactions: dykuma
Suggest using the simplified Stirling formula without the ## \sqrt{2 \pi n} ##.Then the result follows immediately with ## ln(n+x-1)=ln(n) ## in the limit that n approaches infinity... editing...You can include the ## \sqrt{2 \pi n} ## but in the limit that n approaches infinity, this puts the same factor in the numerator that it does in the denominator.
 
Last edited:
Thanks to both of you! The key to solving that problem noticing the what Steven showed, that
stevendaryl said:
ex=limn→∞(1+xn)n

Then the entire problem simplifies quite nicely.
 

Similar threads

  • · Replies 12 ·
Replies
12
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
Replies
12
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K