Is X1 Given S=s a Binomial Distribution in Poisson Variables?

Click For Summary

Discussion Overview

The discussion centers on whether the conditional distribution of a Poisson variable X1 given the sum S = X1 + X2 (where X2 is another independent Poisson variable) follows a binomial distribution. Participants explore the implications of the parameters involved and the nature of the distributions.

Discussion Character

  • Debate/contested, Technical explanation, Exploratory

Main Points Raised

  • One participant states that S is a Poisson variable with mean μ1 + μ2 but is uncertain if X1 given S=s is a binomial distribution.
  • Another participant argues that S cannot be binomial because X2 can take on all integer values, implying that S also ranges over all integers, while a binomial distribution has a finite range.
  • A later reply clarifies that if considering the conditional distribution of X1 given S=s, then X1 cannot range over all integer values, suggesting a different interpretation.
  • Some participants propose a formula for the conditional probability P(X1=x|S=s) that resembles a binomial distribution, but there is acknowledgment of a potential error in the expression regarding coefficients.
  • There is a recognition of misreading the question, with participants revisiting their earlier claims about the distribution type.

Areas of Agreement / Disagreement

Participants express disagreement regarding whether the conditional distribution is binomial, with some supporting the idea while others contest it based on the properties of the distributions involved. The discussion remains unresolved.

Contextual Notes

There are limitations regarding the assumptions made about the ranges of the variables and the implications of conditional distributions. Some mathematical expressions presented may contain inaccuracies that have not been fully clarified.

johnnytzf
Messages
3
Reaction score
0
let X1 and X2 be independent Poisson variables with respective parameters μ1 and μ2. Let S = X1 + X2. Is X1 given S=s a binomial dsitribution? What is the parameters?


I just can show that S is a Poisson with mean μ1 + μ2. But I am not confirm X1 given S is a binomial or not? Someone please help to prove it.
 
Physics news on Phys.org
S cannot be binomial. X2 ranges over all integers. Therefore S will also, while binomial range is finite.
 
If the conditional distribution is intended you want to know about X1 when S = X1 + X2 = s, a fixed value. In this setting X1 cannot range over all integer values.
 
statdad said:
If the conditional distribution is intended you want to know about X1 when S = X1 + X2 = s, a fixed value. In this setting X1 cannot range over all integer values.

You're right. I misread the question.

It looks binomial:

P(X1=x|S=s) = (μ1xμ2(s-x))/{x!(s-x)!(μ1+μ2)s}
 
Last edited:
mathman said:
You're right. I misread the question.

It looks binomial:

P(X1=x|S=s) = (μ1xμ2(s-x))/{x!(s-x)!(μ1+μ2)s}

I was sloppy. There should be a coefficient of s!
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 12 ·
Replies
12
Views
5K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
6K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
3
Views
5K