1 Definition
One of the most frequently used and most intuitive distributions in Exam P.
The easiest way to understand Binomial distribution is to think about tossing coins.
You toss the same coin n times and you want to know how many times you
can get heads.
For each toss, there are only two outcomes -- heads or tails.
The probability of getting heads, represented by p, doesnt change from one
toss to another.
In addition, whether you get a head in one toss doesnt affect whether you get
heads in the next toss (so the result of each toss is independent).
If X = number of heads you get in n tosses of the coin, then X has a binomial
distribution
f ( x )=C n p q
n x
n!
x n x
p q ; x=0,1,2, ,
x ! ( nx ) !
x=0
x=0
F ( x )= C nx p x q nx =
n!
p x q nx
x ! ( nx ) !
()
(n ) ( p e ) ( 1 p )
x
e tx n p x qn x
x
x=0
t x
x=0
[ p et + ( 1 p ) ]
n x
M x ' (0)
n [ p e0 + ( 1 p ) ]
n1
( p e0 )
n ( p+ 1 p ) p
np
n ( n1 ) [ p e + (1 p ) ]
n2
( p e t ) + n [ p e 0 + ( 1 p ) ]
n1
( p e0 )
np ( 1 p+np )
V (X )
2
E ( X 2 ) E( X )2
np ( 1 p+np )np 2
npnp2 +n2 p 2np 2
np(1p)
npq
P ( X 3)
f ( 0 ) + f ( 1 )+ f ( 2 ) +f ( 3 )
0
250
251
252
+C 25 (0.05) (0.95)
+0.093
0.966
Probability of having at least 3 crashes in a day is:
P ( X 3)
1P ( X 2 )
1[f ( 0 ) +f ( 1 ) +f ( 2 ) ]
1[ C 025 (0.05)0 (0.95)250 + C 125 (0.05)1 (0.95)251+ C225 (0.05)2 (0.95)252 ]
0.127
Probability of having more than 3 but less than 6 crashes in a day is:
f ( 4)+ f (5 )
254
255
+C 25 (0.05) (0.95)
0.033
Example 2
A factory has 25 machines working separately, of which 15 are of Type A and 10 of Type
B. On a given day, each machine Type A has a 5% chance of malfunctioning,
independent of the performance of any other machines; each machine Type B has an 8%
chance of malfunctioning, independent of the performance of any other machines.
Let Y represent, on a given day, the total number of machines malfunctioning.
Find P(Y = 2) , the probability of having exactly 2 machines malfunctioning on a given
day.
Solution:
On a given day, the number of Type A machines malfunctioning A X and the number of
Type B machines malfunctioning B X are two independent binomial random variables.
A X is binomial with n =15 and p =5%; B X is binomial with n =10 and p =8% , Y = XA + XB =.
There are only 3 combinations of A X and B X to make Y = 2 : two Type
As malfunction and no Type Bs malfunction; one Type A and one Type B malfunction;
and no Type As malfunction and two Type Bs malfunction. In other words,
150
152
C1 5 (0.08) (0.92)
=0.0685
P ( Y =2 )=0.0585+0.1382+0.0685=0.2652
Example 3
A service guarantee covers 20 TV sets. Each year, each set has 5% chance of failing.
These probabilities are independent.
If a set fails, it is immediately replaced with a new set at the end of the year of failure.
This new set is included in the service guarantee.
Calculate the probability of no more than one failure in the first two years.
Solution:
There are three ways to have zero or one failures during the first two years:
Yr 1
0
Yr 2
0
Prob(Yr 1)
0
20
0
20
200
C (0.05) (0.95)
0.9520
0.9520
( 0.9520 ) =0.1285
20(0.5)0.9519
Tot
al
Total Probability
Prob(Yr 2)
200
C (0.05) (0.95)
0.95
3
C=A B
0.9520
0.3991
The probability of no more than one failure in the first two years is 0.3991