Probability MT22, Branching processes


Flashcards

Definitions

@Define the “offspring distribution” of a branching process.

The distribution for the number of children of each individual.

What assumption is made about individuals reproducing in a branching process, other than having the same offspring distribution?

They reproduce independently.

Generating functions

If the offspring distribution has probability generating function $G(s)$, what is the probability generating function for the number of individuals in the $n$-th generation?

\[G(G(G(G(G\ldots(s)))))\]

repeated $n$ times.

What’s the basic gist of proving the fact $G _ {n+1}(s) = G(G _ n(s))$ for a branching process?

The random sums theorem.

Key quantities

If each individual in a branching process has a mean of $\mu$ children, what is the expected number of individuals in the $n$-th generation?

\[\mu^n\]

What’s the probability that a branching process with generating function $G(s)$ dies out?

The smallest solution to $s = G(s)$ in $[0, 1]$.

Proofs

@Prove that if a branching process has offspring distribution given by probability generating function $G$ then the probability generating function for the number of individuals at generation $n$ (∆generation-n-generating-function) is given by

\[G^n(s)\]

@todo (probability, page 42).

Let $X _ n$ be the number of children in the $n$-th generation of a branching process, and suppose the mean number of children of a single individual is $\mu$. @Prove that (cf. ∆expected-generation-size)

\[\mathbb E[X _ n] = \mu^n\]

@todo