In a Galton-Watson branching process the extinction probability is sometimes given by $$\lim_{t \rightarrow \infty} P[X(t)=0]$$ and sometimes as $$ P[\lim_{t \rightarrow \infty}X(t)=0]$$ Is there a difference between These two formulations? Which is the "correct" one, if yes?

Could you give me a link to a paper where this is written?

Here is the Wikipedia page, but unfortunately, the linked papers there do not help..

Thank you for your help

  • $\begingroup$ In general, there is a difference between the two formulations. Unfortunately, I'm not familiar with Galton-Watson processes... $\endgroup$ – saz Aug 16 '14 at 10:16
  • $\begingroup$ Literally, the former means the limit of a sequence of real numbers; while the latter means the probability of the convergence of a given sequence. So there should be a difference. $\endgroup$ – Megadeth Aug 16 '14 at 10:18
  • $\begingroup$ I see that this Argument implies that it can't be the same exactly, as These are two different formulations, the one a squence of random variables, the other a sequence of probabilities. What I wonder whether the two Limits are then the same or not.. Does this also hold, that this is in General not the same? $\endgroup$ – user146358 Aug 16 '14 at 10:27

In the context of branching processes, the two statements are indeed equivalent. This is due to the following facts:

  • The value of each random variable $X(t)$ is almost surely a nonnegative integer.
  • The events $[X(t)=0]$ are nondecreasing with respect to $t$.

Thus, $P(X(t)=0)\to P(A)$ when $t\to\infty$, where $A=\bigcup\limits_t[X(t)=0]$. On the other hand, the event $[\lim\limits_{t\to\infty}X(t)=0]$ is also $[\exists t,X(t)=0]=A$, QED.

To sum up, the reason why the assertion holds is that, if $x:t\mapsto x(t)$ is an integer valued function such that if $x(t)=0$ then $x(s)=0$ for every $s\geqslant t$, then $x(t)\to0$ when $t\to+\infty$ if and only if $x(t)=0$ for some $t$.

  • $\begingroup$ thank you for your answer. that helps a lot. Do you know a link to a paper, where the mathematical Definition of a Galton-Watson process is stated? All I can find is Wikipedia and the references there do not lead to formal definitions, either. $\endgroup$ – user146358 Aug 16 '14 at 12:57
  • 1
    $\begingroup$ The WP page has a formal definition. Some recent introductory notes on the subject called "Probability on Trees: An Introductory Climb" are available on Yuval Peres webpage. $\endgroup$ – Did Aug 16 '14 at 13:03
  • $\begingroup$ I know that Wikipedia has a formal Definition, and this is enough for me. But the Problem is, that I am not allowed to reference to Wikipedia. Thank you. $\endgroup$ – user146358 Aug 16 '14 at 13:05
  • $\begingroup$ Quote: "the references there do not lead to formal definitions". $\endgroup$ – Did Aug 17 '14 at 9:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.