Professional Documents
Culture Documents
Binomial PDF
Binomial PDF
b(x; n, p) = n
x
px (1 − p)n−x
This is the probability of having x successes in a series of n independent trials when the
probability of success in any one of the trials is p. If X is a random variable with this
probability distribution,
n
X x
E(X) = x n
x
p (1 − p)n−x
x=0
Xn
= n!
x x!(n−x)! px (1 − p)n−x
x=0
Xn
= (x−1)!(n−x)! p (1
n! x
− p)n−x
x=1
since the x = 0 term vanishes. Let y = x −1 and m = n −1. Subbing x = y +1 and n = m+1
into the last sum (and using the fact that the limits x = 1 and x = n correspond to y = 0
and y = n − 1 = m, respectively)
m
(m+1)! y+1
X
E(X) = y!(m−y)!
p (1 − p)m−y
y=0
m
X
= (m + 1)p y!(m−y)! p (1
m! y
− p)m−y
y=0
m
X
= np y!(m−y)! p (1
m! y
− p)m−y
y=0
Setting a = p and b = 1 − p
m
X m
X
y!(m−y)! p (1 − p)m−y = = (a + b)m = (p + 1 − p)m = 1
m! y m! y m−y
y!(m−y)! a b
y=0 y=0
c Joel Feldman. 2000. All rights reserved. 1
so that
E(X) = np
= n(n − 1)p2
So the variance of X is
= np(1 − p)
c Joel Feldman. 2000. All rights reserved. 2