This preview shows page 1. Sign up to view the full content.
Unformatted text preview: e is the binomial(N, x/N ) distribution, i.e., the number of
successes in N trials when success has probability x/N , so the mean number of
successes is x. From this it follows that if we deﬁne h(x) = x/N , then
p(x, y )h(y )
y Taking a = N and b = 0, we have h(a) = 1 and h(b) = 0. Since Px (Va ^ Vb <
1) > 0 for all 0 < x < N , it follows from Lemma 1.27 that
Px (VN < V0 ) = x/N (1.18) i.e., the probability of ﬁxation to all A’s is equal to the fraction of the genes
that are A.
Our next topic is non-fair games.
Example 1.43. Gambler’s ruin. Consider a gambling game in which on
any turn you win $1 with probability p 6= 1/2 or lose $1 with probability 1 p.
Suppose further that you will quit playing if your fortune reaches $N . Of course,
if your fortune reaches $0, then the casino makes you stop. Let
h(x) = Px (VN < V0 )
be the happy event that our gambler reaches the goal of $N before going
bankrupt when starting with $x. Thanks to our deﬁnition of Vx as the minimum of n 0 with Xn = x we have h(0) = 0, and h(N...
View Full Document
This document was uploaded on 03/06/2014 for the course MATH 4740 at Cornell University (Engineering School).
- Spring '10
- The Land