Generating functions in the branching process.

  • Thread starter Thread starter stukbv
  • Start date Start date
  • Tags Tags
    Functions Process
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
1 reply · 2K views
stukbv
Messages
112
Reaction score
0

Homework Statement



I am told that I have particles which each yield a random number of offspring of known distribution independently from each other and from the past generations.

Xn is the number of particles in the nth generation
The distribution of a typical family size is Z and so Xn is the sum of appropriate Zi's

I need a generating function of the number Xn in the nth generation.


2. The attempt at a solution

I know that Fn+1(s) = E [sXn+1]
from the definition of generating functions and how to derive them.

But my lecturer then goes on to say that = ƩE[sXn+1|Xn=j] * P[Xn=j ]

Summed over j.

How does he get from one to the other? If I can make this link then I can go on to show what I need to!

Thank you!
 
Physics news on Phys.org
stukbv said:

Homework Statement



I am told that I have particles which each yield a random number of offspring of known distribution independently from each other and from the past generations.

Xn is the number of particles in the nth generation
The distribution of a typical family size is Z and so Xn is the sum of appropriate Zi's

I need a generating function of the number Xn in the nth generation.2. The attempt at a solution

I know that Fn+1(s) = E [sXn+1]
from the definition of generating functions and how to derive them.

But my lecturer then goes on to say that = ƩE[sXn+1|Xn=j] * P[Xn=j ]

Summed over j.

How does he get from one to the other? If I can make this link then I can go on to show what I need to!

Thank you!

It's just a standard result in Probability. Suppose [itex]\{A_k \}[/itex] is a partition of the sample space [itex]\Omega[/itex], meaning that the A_k are disjoint and their union is Ω. Then, for any discrete random variable B we have [tex]\Pr \{B=j\} = \sum_k \Pr\{B=j|A_k\} \Pr \{A_k\}.[/tex] Thus, for any f >= 0 we have
[tex]E f(B) = \sum_j f(j) \Pr\{B=j \} = \sum_k \Pr\{A_k\} \sum_j f(j) \Pr\{B=j|A_k\}<br /> =\sum_k E[f(B)|A_k] \Pr\{A_k\},[/tex]
where I have swapped the order of summation, which is OK for a positive function.

RGV
 
Last edited: