ch02 JWBK035-Vince February 12, 2007 6:50 Char Count= 0
78 HANDBOOK OF PORTFOLIO MATHEMATICS
. 2499 /. 00001111 =N− 1
. 2499 /. 00001111 + 1 =N
22 , 491 + 1 =N
N= 22 , 492
Thus, we need to witness a 51% win rate over 22,492 trials to be 99.865%
certain that we will see at least 51% wins.
The Geometric Distribution
Like the Binomial, theGeometric Distribution, also a discrete distribution,
occurs as a result of N independent Bernoulli trials. The Geometric Distri-
bution measures the number of trials before the first success (or failure).
The probability density function, N′(X), is:
N′(X)=Q(X−1)∗P (2.39)
where: P=The probability of success for a given trial.
Q=The probability of failure for a given trial.
In other words, N′(X) here measures the number of trials until the first
success. The cumulative density function for the Geometric is therefore:
N(X)=
∑X
J= 1
Q(J−1)∗P (2.40)
where: P=The probability of success for a given trial.
Q=The probability of failure for a given trial.
Figures 2.22 and 2.23 illustrate the probability density and cumulative
probability ability (i.e., cdf), respectively, of the Geometric Distribution.
Other properties of the Geometric are:
Mean= 1 /P (2.41)
Variance=Q/P^2 (2.42)
where: P=The probability of success for a given trial.
Q=The probability of failure for a given trial.