Proving the Limit of Binomial Distribution as Poisson: Formal Proof

  • Thread starter Thread starter relinquished™
  • Start date Start date
  • Tags Tags
    Limit
Click For Summary
SUMMARY

The limit of the binomial distribution approaches the Poisson distribution as n approaches infinity. Specifically, the expression \(\lim_{n\to +\infty} \frac{n!}{(n-x)! (n-k)^x}\) simplifies to show that it approaches 1. This is achieved by rewriting the expression as \(\frac{n(n-1)(n-2)...(n-x+1)}{(n-k)^x}\) and analyzing the limits of the resulting fractions. The proof confirms that both components of the expression converge appropriately, establishing the relationship definitively.

PREREQUISITES
  • Understanding of binomial distributions and their properties
  • Familiarity with Poisson distributions
  • Knowledge of limits in calculus
  • Experience with factorial notation and simplification
NEXT STEPS
  • Study the derivation of the Poisson distribution from the binomial distribution
  • Learn about the properties of limits in calculus
  • Explore applications of the Poisson distribution in statistics
  • Investigate the relationship between large sample sizes and distribution convergence
USEFUL FOR

Mathematicians, statisticians, and students studying probability theory who seek to understand the convergence of distributions and formal proofs in statistical contexts.

relinquished™
Messages
79
Reaction score
0
I'm a little stuck in my proof here. As I was trying to prove that the limit of a binomial distribution is the poisson distribution, I encountered this:

[tex] <br /> \lim_{n\to +\infty} \frac{n!}{(n-x)! (n-k)^x}<br /> [/tex]

where x and k are arbitrary constants.


The books say that this approaches 1, but shows no formal proof. How are we sure that this approaches 1 as n gets larger? In short, what's the formal proof?

Thanx for any help
 
Physics news on Phys.org
Let's see... you have n things multiplied together on the top (1, 2, 3, ..., n), and you have n things multipled together on the bottom: (n - x) of them in (n-x)! and x of them in (n-k)^x. My first instinct would be to try and group terms in the numerator with terms in the denominator.
 



To prove that the limit of the binomial distribution is the Poisson distribution, we need to show that the expression \frac{n!}{(n-x)! (n-k)^x} approaches 1 as n approaches infinity. This can be done by using the definition of a limit.

First, let's rewrite the expression as \frac{n(n-1)(n-2)...(n-x+1)}{(n-k)^x}. We can then rewrite this as \frac{n^x}{(n-k)^x} \cdot \frac{(n-1)(n-2)...(n-x+1)}{n(n-1)(n-2)...(n-x+1)}. Notice that the first fraction approaches 1 as n approaches infinity, since the numerator and denominator both have a factor of n that cancels out. So, we can focus on the second fraction.

Using the property of limits, we can rewrite the second fraction as \frac{(n-1)x}{n(n-1)(n-2)...(n-x+1)} \cdot \frac{(n-2)...(n-x+1)}{(n-1)x}. Now, as n approaches infinity, we can see that the first fraction approaches 0, since the numerator is a constant (n-1)x and the denominator is increasing without bound. The second fraction also approaches 1, since the numerator and denominator both have a factor of n that cancels out.

Therefore, the entire expression approaches 1 as n approaches infinity. This proves that the limit of the binomial distribution is the Poisson distribution.

I hope this helps clarify the formal proof for you. If you have any further questions, don't hesitate to ask.
 

Similar threads

  • · Replies 21 ·
Replies
21
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K
Replies
8
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 5 ·
Replies
5
Views
2K
Replies
17
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K