What is the distribution of A given B and B's distribution?

  • Thread starter Thread starter spitz
  • Start date Start date
  • Tags Tags
    Distribution Fun
Click For Summary

Homework Help Overview

The discussion revolves around finding the distribution of a random variable A given another random variable B, where A follows a Poisson distribution conditioned on B, and B follows an Exponential distribution. Participants are exploring the relationship between these distributions and the marginal distribution of A.

Discussion Character

  • Exploratory, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants are attempting to derive the marginal distribution of A by using the law of total expectation and integrating the joint probability density function of A and B. There are discussions about using integration by parts and the Gamma function to simplify the calculations.

Discussion Status

Some participants have made progress in their calculations and are questioning the next steps, particularly regarding the evaluation of integrals and the potential identification of the resulting distribution as geometric. There is no explicit consensus on the final form of the distribution yet.

Contextual Notes

Participants are working under the constraints of the problem setup, specifically the nature of A being discrete and B being continuous, and the need to find the marginal distribution of A from the joint distribution.

spitz
Messages
57
Reaction score
0

Homework Statement



I need to find the distribution of [itex]A[/itex]

Homework Equations



[itex]A|B\sim \text{Poisson}(B)[/itex]

[itex]B\sim \text{Exponential}(\mu)[/itex]

The Attempt at a Solution


[itex]\displaystyle P(A=k)=E[P(A=k|B)]=E\left[e^{-B}\cdot\frac{B^k}{k!}\right]=\ldots[/itex]

Not sure how to calculate this...
 
Physics news on Phys.org
spitz said:

Homework Statement



I need to find the distribution of [itex]A[/itex]

Homework Equations



[itex]A|B\sim \text{Poisson}(B)[/itex]

[itex]B\sim \text{Exponential}(\mu)[/itex]

The Attempt at a Solution


[itex]\displaystyle P(A=k)=E[P(A=k|B)]=E\left[e^{-B}\cdot\frac{B^k}{k!}\right]=\ldots[/itex]

Not sure how to calculate this...

Let me guess the A|B is Poisson with parameter B, where B is exponential with parameter μ, so A is discrete and B is continuous, the problem is finding the marginal distribution of the discrete A, the joint pdf of A and B is f(A,B)=Poisson(A;B)*Exp(B;μ), so that the marginal pdf of A is ∫dB f(A,B)=∫dB Poisson(A;B)*Exp(B;μ);
If it ends up becoming evaluating [itex]\int dB e^{-B} B^k[/itex] for some k, start with k=0, and try integration by parts to come up with a recursive relation from k to k+1, and evaluate in closed form if possible.
 
Last edited:
[tex]=\displaystyle\int_{0}^{\infty}e^{-b}\frac{b^k}{k!}\mu e^{-\mu b}db[/tex]

[tex]=\frac{\mu}{k!}\displaystyle\int_{0}^{\infty}b^ke^{-b(1+\mu)} db[/tex]

This is where I am at... not sure what to do now.
 
spitz said:
[tex]=\displaystyle\int_{0}^{\infty}e^{-b}\frac{b^k}{k!}\mu e^{-\mu b}db[/tex]

[tex]=\frac{\mu}{k!}\displaystyle\int_{0}^{\infty}b^ke^{-b(1+\mu)} db[/tex]

This is where I am at... not sure what to do now.

Integration by parts to go from k to k+1
 
spitz said:
[tex]=\displaystyle\int_{0}^{\infty}e^{-b}\frac{b^k}{k!}\mu e^{-\mu b}db[/tex]

[tex]=\frac{\mu}{k!}\displaystyle\int_{0}^{\infty}b^ke^{-b(1+\mu)} db[/tex]

This is where I am at... not sure what to do now.

Change variables in the integral; you ought to be able to get a Gamma function; seehttp://en.wikipedia.org/wiki/Gamma_function .

RGV
 
[itex]= \frac{\mu}{k!}\frac{\Gamma(k+1)}{(1+\mu)^{k+1}}= \frac{\mu}{(1+\mu)^{k+1}}[/itex]

[itex]= \frac{\mu}{1+\mu}\left(1-\frac{\mu}{1+\mu}\right)^k[/itex]

Would this be geometric then? What is the parameter?
 
Last edited:

Similar threads

Replies
2
Views
1K
Replies
6
Views
2K
Replies
6
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 15 ·
Replies
15
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 8 ·
Replies
8
Views
3K