Module 2 Chapter 3 solutions

Chapter 3 solutions

Solution 3.1

The random variable X has a discrete uniform distribution, taking the values 1,2,...,10, so
k  10 .
The mean and variance of this distribution are:
k  1 11
E( X )    5.5
2 2

k 2  1 99
var( X )    8.25
12 12

Solution 3.2

Let X be the number of visits (out of the next 10 visits) that result in a referral. Then X
follows a binomial distribution with n  10 and p  0.28 , and the required probability is:

P  X  4   10C4  0.284  0.726  0.1798

Solution 3.3
The probability that a report is made is:
P  more than one injury   0.23  0.17  0.09  0.04  0.53

Let X be the number of reports made in the next 20 games. Then X follows a binomial
distribution with n  20 and p  0.53 .

The expected number of reports is:

E  X   np  20  0.53  10.6

The standard deviation of the number of reports is:

npq  20  0.53  0.47  2.232

Solution 3.4

The number of claims X follows a binomial distribution with n  10 and p  0.125 .

The expected number of claims is:

E  X   np  10  0.125  1.25

The variance is:

var  X   npq  10  0.125  0.875  1.09375

The standard deviation is:

var  X   1.09375  1.04583

1
Chapter 3 solutions Module 2

Since X is the number of claims and 10X is the payment (in $million), the reserve is:
10  1.25  2  1.04583   33.417 ($million)

Solution 3.5
The general form of the moment generating function of a binomial random variable is:

 
n
M X (t )  pet  q

so in this case, p  0.6, q  0.4 and n  20 . Therefore:

E ( X )  np  12

var( X )  npq  4.8

Solution 3.6

Let X be the number of accidents at the road junction in the next year. Then X follows a
Poisson distribution with   5 , and the required probability is:

e   3 e 5 53
P  X  3    0.1404
3! 3!

Solution 3.7

Let X be the number of cyclones in a particular year. Then X follows a Poisson distribution
with   2.8 , and we have:

P ( X  1)  1  P  X  0   1  e    1  e 2.8  1  0.06081  0.93919

P  X  3   1  P  X  0   P  X  1  P  X  2 

e 2.8  2.8 e 2.8  2.82

 1  e 2.8  
1! 2!
 1  0.060810  0.170268  0.238375  0.53055
Hence, the required probability is:
P  X  3  X  1 P  X  3 0.53055

P X 3 X 1   P  X  1

P  X  1

0.93919
 0.5649

Solution 3.8
Given that:
P  X  0   2  P  X  1
e  
e   2 
1!
it follows that:
1
  0.5
2
The variance of the distribution is   0.5 .

2
Module 2 Chapter 3 solutions

Solution 3.9
The cumulant generating function of the given Poisson distribution is:

C X (t )  ln M X  t   16(et  1)  16et  16

Differentiating once with respect to t , we have:

C X  t   16et

Differentiating again, we have:

C X  t   16et

Differentiating again, we have:

C X  t   16et

Setting t  0 , we determine the variance and skewness as:

var( X )  C X  0   16

skew ( X )  C X  0   16

So, the coefficient of skewness is:

skew ( X ) 16 1 1
    0.25
1.5 1.5 0.5 16
 var( X ) 16 16

Solution 3.10

(i) The probability that there are at least 4 injuries in a game is 0.09  0.04  0.13 .
Let X be the number of games until the first game in which there are at least 4
injuries. Then X follows a geometric distribution with p  0.13 . Hence:

P  X  6   0.875  0.13  0.0648

(ii) The mean of this geometric distribution is:

1 1
EX    7.6923
p 0.13

Solution 3.11
The probability of success (ie all 5 coins are heads) on a single trial is:
5
 1 1
 2   32
 
Then N (the number of trials until all five coins are heads) follows a geometric distribution
1
with p  .
32
We’re asked for the probability that it takes 20 or more trials until the first success. We can
use the cumulative distribution function to calculate the probability that it takes 19 or fewer
trials, where:

F (n )  P  N  n   1  q n

3
Chapter 3 solutions Module 2

Hence:
19

P  N  20   1  P  N  19   1  1  q19  q19   
 31 

 32 
 0.5470

Solution 3.12

Let X be the number of records until the 100th patient is reached. X follows a negative
binomial distribution with parameters p  0.05 and k  100 . The expected number of
records that the researcher will have to look through is the mean of this distribution, where:
k 100
EX    2,000
p 0.05

Solution 3.13

Let X be the number of months until the third month in which at least one accident occurs.
Then X follows a negative binomial distribution with p  0.3 and k  3 .

We want the probability that the third month in which at least one accident occurs is the sixth
month, ie P  X  6  . The required probability is therefore:

P  X  6   5C2 0.730.33  0.09261

Solution 3.14
n
Let S   Xi where each X i has a negative binomial distribution with parameters ki and p
i 1
and a moment generating function given by:
ki
 pet 
M Xi t    
 1  qet 
 
Then:
MS  t   M X1  t   M X 2  t     M X n  t 

i
ki
 pet 1
 
 1  qet 
 
We can see that this moment generating function is the moment generating function of a
negative binomial distribution with parameters  ki and p . By the uniqueness property of
moment generating functions, we can therefore conclude that if n independent X i random
variables each have a negative binomial distribution with parameters ki and p , then  Xi
follows a negative binomial distribution with parameters  ki and p .