STAT/MA 416000

PRACTICE QUESTION SET 5(WITH KEY)

(1) A discrete random variable N has the following probability function:




 p, x=1


f (x) = 2p, x=3




1 − 3p, x = 5

You are also give E[N ] = 3.8. Determine p.

E[N ] = p + 6p + 5 − 15p = 3.8

3
p=
20

(2) The length of a staff meeting, in hours, is a random variable having the following probability density
function: 
5(x − 0.5)4 , 0.5 ≤ x ≤ 1.5

f (x) =
0,

otherwise

Calculate the expected length of a staff meeting.

Z 1.5 Z 1.5
f (x)dx = 5(x − 0.5)4 dx
0.5 0.5
1.5
= (x − 0.5)5 

0.5

=1
Therefore the probability at F (0.5) = F (1.5) = 0, hence
Z 1.5
E(X) = f (x) · xdx
0.5
Z 1.5
= 5x(x − 0.5)4 dx
0.5
1
= 1.5 −
6
4
=
3

Date: October 4, 2019.

1
(3) Let X be a continuous random variable with density function

θx + 3 θ3/2 x2 , for 0 < x <
 √1
2 θ
f (x) =
0,

otherwise

where θ > 0. What is the expected value of X?

Z √1 Z √1
θ θ 3
f (x)dx = θx + θ3/2 x2 dx
0 0 2
=1
Hence 1
Z √
θ
E(X) = f (x)xdx
0
Z √1
θ 3
= θx2 + θ3/2 x3 dx
0 2
17
= √
24 θ

(4) Let X be a continuous random variable with density function.


 |x| , for − 2 ≤ x ≤ 4

10
f (x) =
0,

otherwise

Calculate the expected value of X.

Z 4
E(X) = f (x)xdx
−2
0 4
x2 x2
Z Z
= − dx + dx
−2 10 0 10
28
=
15

(5) Annual losses on an insurance policy have a distribution with probability density function


−x2 /200
e √

 ,x > 0
f (x) = 5 2π

0,

otherwise

Calculate the expected value of X.

2
 ∞ ∞ 2
e−x /200
Z Z
f (x)dx = √ dx
0 0 5 2π
Z ∞
1 2
= √ e−x /200 dx
5 2π 0
=1
Hence we can calculate the expectation
Z ∞
1 2
E(X) = √ e−x /200 xdx
5 2π 0
1
= √ · 100
5 2π
10
=√
π

(6) The distribution of loss sizes X on an insurance policy has the following density function:

 2000 , x > 2000

f (x) = x2
0,

otherwise

An insurance policy has a policy limit of 5000. Calculate the expected payment per loss on this
policy.

For x ∈ (2000, 5000), take integration

Z 5000
F (5000) = f (x)dx
2000
Z 5000
2000
= dx
2000 x2
2
=1−
5
3
=
5
Therefore, p(5000) = 25 . Hence we can calculate the expectation
Z 5000
2000
E(X) = dx + 2000
2000 x
5
= 2000 ln + 2000
2

(7) An insurance company’s monthly claims are modeled by a continuous positive random variable X,
whose probability density function is proportional to (1 + x)−4 , for 0 < x < ∞. Calculate the
company’s expected monthly claims.

3
 Integrate over (0, ∞), we get
Z ∞ Z ∞
f (x)dx = c(1 + x)−4 dx
0 0
c
=
3
Therefore, c = 3, hence the expected monthly claim is
Z ∞
E(X) = 3x(1 + x)−4 dx
0
Z ∞ Z ∞
= 3(1 + x)−3 dx − 3(1 + x)−4 dx
0 0
3
= −1
2
1
=
2

(8) Let X be a continuous random variable with density function


p − 1, x > 1

f (x) = xp
0, otherwise


Calculate the value of p such that E[X] = 2.

Integrate over (1, ∞), we get

∞ ∞
p−1
Z Z
f (x)dx = dx
1 1 xp
∞
1−p 

= (−x )
1

=1

Therefore, p > 1, hence the expectation is

∞
p−1
Z
E(X) = x dx
1 xp
∞
p−1
Z
= dx
1 xp−1
∞
p − 1 2−p 
=( x )
2−p 1
p−1
=
p−2
Hence p = 3. where in the last line we use p > 2 (Because p = 2 and 2 > p > 1 will lead to
infinite expectations).

4
 (9) A manufacturer’s annual losses follow a distribution with density function
2.5

 2.5(0.6) , x > 0.6

x 3.5
f (x) =
0,

otherwise

To cover its losses, the manufacturer purchases an insurance policy with an annual deductible of 2.
Calculate the mean of the manufacturer’s annual losses not paid by the insurance policy.

We will integrate min(X, 2) over the density function. We break the integral up into the
component below 2 and the component above 2. For the component above 2, we integrate
2f (x):

∞ ∞
5(0.6)2.5 0.62.5 
Z
dx = − 2.5  = 0.098590
2 x3.5 x 2
For the component below 2, we integrate xf (x):
Z 2
2.5(0.6)2.5
dx = 0.835683
0.6 x2.5
Hence the expectation is 0.098590 + 0.835683 = 0.9343

(10) An insurance policy reimburses a loss up to a benefit limit of 10. The policyholder’s loss, Y , follows
a distribution with density function:

2y −3 , y > 1

f (y) =
0, otherwise


Calculate the expected value of the benefit paid under the insurance policy.

We will integrate min(Y, 10) over the density function. We break the integral up into the
component below 10 and the component above 10. For the component above 10, we integrate
10f (x):

Z ∞ ∞
20 10 
3
dy = − 2  = 0.1
10 y y 10
For the component below 10, we integrate xf (x):
Z 10 10
2 2 
dy = −  = 2 − 0.2 = 1.8
1 y2 y 1
Hence the expectation is 1.8 + 0.1 = 1.9

(11) Let X be a continuous random variable with density function


 1 x(1 + 3x), for 1 < x < 3

30
f (x) =
0,

otherwise
5
 Calculate E[1/X].

Note that Z 3 Z 3
1
f (x)dx = x(1 + 3x)dx
1 1 30
=1
Therefore Z 3
1 1
E[1/X] = · x(1 + 3x)dx
1 x 30
Z 3
1
= (1 + 3x)dx
1 30
3
1 3 2 
= (x + x )
30 2 1
7
=
15

(12) You are given that E[X] = 10 and Var(X) = 40. Calculate E[(X + 5)2 ].

E[(X + 5)2 ] = E[X 2 ] + 10E[X] + 25

= Var(X) + E[X]2 + 10E[X] + 25

= 40 + 100 + 100 + 25

= 265

(13) Let X be a random variable such that E[X] = 2, E[X 3 ] = 9, and E[(X − 2)3 ] = 0. What is Var(X)?

We first calculate E[X 2 ]. Since E[(X −2)3 ] = E[X 3 −6X 2 +12X −8] = 0, hence E[X 2 ] = 25/6

Var(X) = E[X 2 ] − E[X]2

25
= −4
6
1
=
6

(14) A recent study indicates that the annual cost of maintaining and repairing a car in a town in Ontario
averages 200 with a variance of 260. A tax of 20% is introduced on all items associated with the
maintenance and repair of cars(i.e., everything is made 20% more expensive). Calculate the variance
of the annual cost of maintaining and repairing a car after the tax is introduced.

We know that E[X] = 200, Var(X) = 260,

Var(1.2X) = 1.44 · Var(X) = 374.4

6
(15) A random variable X has the cumulative distribution function.



0,

 x<1

 2
x − 2x + 2
F (x) = ,1 ≤ x < 2


 2

1, x≥2


Calculate the variance of X.

F (1) = 1/2, so we have a mixed distribution here, However, F (2− ) = F (2) = 1. The density
function between 1 and 2 is
f (x) = x − 1
R2
When calculating k th moments, we must add 1/2(1k ) = 1/2 to 1
xk f (x)dx to account for
the point mass at 1. The moments of X are
Z 2
E[X] = 1/2 + x(x − 1)dx
1
5 1
= +
6 2
4
= ,
3
23 16 5
Var(X) = E[X 2 ] − E[X]2 = − =
12 9 36

(16) The distribution of losses X on an insurance policy has the following probability density function:


 500 , x > 500

f (x) = x2
0, otherwise


The policy has a benefit limit of 2000. Calculate the variance of the payment per loss on the policy.

7
Note that for x ∈ (500, 2000)
Z 2000
F (x) = f (x)dx
500
Z 2000
500
= dx,
500 x2
3
=
4
Therefore this is a mixed distribution and F (2000) = 1/4. Hence the moments are.
Z 2000
E[X] = xf (x)dx + 500
500

= 500 ln 4 + 500
Z 2000
E[X 2 ] = x2 f (x)dx + 106
500

= 1.75 × 106

Var(X) = 1.75 × 106 − (500 ln 4 + 500)2

= 1.5 × 106 − 2.5 × 105 ln 42 − 5 × 105 ln 4

≈ 326399.81