MATH 2930 Assignment # 3 Solutions [68 marks] Fall 2018

1. In commuting to work, a person must first get on a bus near her home, and then transfer to a second
bus. If the waiting time (in minutes) at each stop has a uniform distribution, then the total waiting
time Y has pdf 
1
y 0≤y<5


 25





f (y) = 2 1
− y 5 ≤ y ≤ 10
 5 25






 0 otherwise

(a) [Total 2 marks] Sketch the graph of the pdf of Y .

Z ∞
(b) [Total 3 marks] Show that f (y)dy = 1.
−∞

(c) [1 mark for the correct probability, 2 mark for the work; total 3 marks] What is the probability the
total waiting time is at most 3 minutes?

(d) [1 mark for the correct probability, 2 marks for the work; total 3 marks] What is the probability the
total waiting time is at most 8 minutes?

(e) [Total 2 marks]What is the probability the total waiting time is between 3 to 8 minutes?
Solution:
(a)

1
 (b)
Z ∞ Z 5 Z 10
f (y)dy = f (y)dy
f (y)dy +
−∞ 0 5
Z 5  Z 10

1 2 1
= ydy + − y dy
0 25 5 5 25
5 10 10
1 2  2  1 2 
= y  + y − y 
50 0 5 5 50 5
1 2 1
= 52 + (10 − 5) − (102 − 52 )
50 5 50
25 25
= +
50 50
= 1.
Z 3 Z 3 Z 3 3
1 1 2  9
(c) P(Y ≤ 3) = f (y)dy = f (y)dy = ydy = y  = = 0.18.
−∞ 0 0 25 50 0 50
Z 8 Z 8 Z 5 Z 8 Z 5 Z 8 
1 2 1
(d) P(Y ≤ 8) = f (y)dy = f (y)dy = f (y)dy + f (y)dy = ydy + − y dy.
−∞ 0 0 5 0 25 5 5 25
Evaluating this integral
5 leads
8 to 8
1 2  2  1  1 2 1 6 64 23
P(Y ≤ 8) = y  + y  − y 2  = 52 + (8 − 5) − (82 − 52 ) = 1 + − = = 0.92
50 0 5 5 50 5 50 5 50 5 50 25
(e) Using cumulative probabilities, we have P(3 ≤ Y ≤ 8) =P (Y ≤ 8)−P (Y ≤ 3) = 0.92−0.18 = 0.74.
Z 8
Alternatively, evaluate the integral directly; that is, P(3 ≤ Y ≤ 8) = f (y)dy = · · ·
3

2. [Total 3 marks] Let X be a continuous random variable with pdf

kθk


 k+1 x≥θ
f (x) = x


0 otherwise
The parameters k and θ are constants (with respect to x). Find an expression for P(X ≤ b) for any
fixed b > θ.
Solution: b
b b b
kθk θk
Z Z Z  
k 1  1 1
 k
P(X ≤ b) = f (x)dx = f (x)dx = dx = −kθ = kθ − + = 1 − .
−∞ θ θ xk+1 kxk θ kbk kθk bk
3. Consider the pdf for the total waiting time Y for two buses,

1
y 0≤y<5


25






f (y) = 2 1
− y 5 ≤ y ≤ 10
5 25







 0 otherwise

Note: this is the same pdf as in Question 1.

2
(a) [1 mark for the correct value of F(y) for each interval; 1 mark for each correct interval, no marks for
the work; total 8 marks] Compute the cumulative distribution function of Y , denoted F (y).

(b) [1 mark for the correct value] Use F (y) (from a) to determine P (4 ≤ Y ≤ 9).

(c) [1 mark for the correct value; 2 marks for the work; total 3 marks] Compute the expected value of Y ;
that is, E[Y ].

(d) [1 mark for the correct value; 2 marks for the work; total 3 marks] Compute the variance of Y ; that
is, V [Y ].
Solution:
(a) Let F (Y ) be the cumulative distribution function of Y . To determine F (Y ), integrate f (y) on
the appropriate intervals. For y < 0, F (y) = 0 since there is no area under the graph of the density
function to the left of y < 0. For y ∈ [0, 5], we have
Z y Z y y
1 1 1 2  y2
F (y) = f (u)du = udu = u  = .
25 ∞ 25 0 50 0 50

For y ∈ [5, 10], we have

Z y
F (y) = f (u)du
∞
Z 5 Z y 
1 2 1
= udu + − u du
25 0 5 5 25
5 y y
1 2  2  1 2 
= u  + u − u 
50 0 5 5 50 5
1 2 1
= 52 + (y − 5) − (y 2 − 52 )
50 5 50
2
y 2
= − + y − 1.
50 5

For y > 10, F (y) = 1 since all the area is accumulated to the left of y > 10.

 0 y<0
y2



0≤y<5


50




F (y) =
y2 2
− + y − 1 5 ≤ y < 10





 50 5



1 10 ≤ y


(b) Using F (y) in (a), we can determine P (4 ≤ Y ≤ 9) as

92 2(9) 42 −81 + 180 − 50 − 16 33

P (4 ≤ Y ≤ 9) = F (9) − F (4) = − + −1− = = .
50 5 50 50 50

3
 (c) The expected value of Y is
Z ∞ Z 5 Z 10
µ = E[Y ] = yf (y)dy = yf (y)dy + yf (y)dy
−∞ 0 5
Z 5 Z 10  
1 2 2 1 2
= y dy + y − y dy
0 25 5 5 25
5 10 10
1  1  1 
= y 3  + y 2  − y 3 
75 0 5 5 75 5
1 1 1 1
= 53 + (10)2 − (5)2 − (10)3 − 53


75 5 5 75
= 5.

Alternatively, by the symmetry of f (y) (see Question 1a), we see the center (that is, the expected
value) is 5.
(d) The variance of Y (using part (b)) is

V [Y ] = E[Y 2 ] − (E[Y ])2

= E[Y 2 ] − µ2
Z ∞
= y 2 f (y)dy − 52
−∞
Z 10
= y 2 f (y)dy − 25
Z0 5 Z 10  
1 3 2 2 1 3
= y dy + y − y dy − 25
0 25 5 5 25
5 10 10
1 4  2 3  1 4 
= y + y − y − 25
100 0 15 5 100 5
1 4 2 1
(10)3 − 53 − (10)4 − 54 − 25



= 5 +
100 15 100
50
= .
12

Alternatively the definition of variance can be used; that is,

Z ∞
2
V [Y ] = E[(Y − µ) ] = (y − µ)2 f (y)dy.
−∞

4. Let Z be a continuous random variable with Z ∼ N (0, 1).

(i) [1 mark for the correct value; no work required; total 1 mark] Find the value of P (Z < −1.00).

(ii) [1 mark for the correct value; 2 marks for the work; total 3 marks] Find the value of P (|Z| < 1.00).
Note | · | denotes the absolute value function.

(iii) [1 mark for the correct value; 2 marks for the work; total 3 marks] Find b such that P (Z > b) = 0.9925.

(iv) [1 mark for the correct value; no work required; total 1 mark] Find the 33th percentile.

4
(v) [1 mark for the correct value; no work required; total 1 mark] Find the value of the critical value z0.025 .
Solution:
(i) From the z-table, we see P (Z < −1.00) = 0.1587.
(ii) From the z-table, we see P (|Z| < 1.00) = P (−1.00 < Z < 1.00) = P (Z < 1.00) − P (Z < −1.00) =
0.8413 − 0.1587 = 0.6826.
(iii) We see P (Z > b) = 1 − P (Z ≤ b) = 0.9925, which implies P (Z ≤ b) = 0.0075 and hence from the
z-table, b = −2.43.
(iv) From the z-table, we see P (Z < z) = 0.33 corresponds to z=-0.44. Therefore, the 33th percentile
is -0.44.
(v) Since the critical value z0.025 indicates an area of 0.025 to the RIGHT of the critical value, then
an area of 1-0.025=0.975 is to the LEFT of the critical value. From the z-table, the z-value associated
to an area of 0.975 is 1.96, and hence z0.025 = 1.96.

5. Let X represent hair length, and assume it is normally distributed. Determine the probability the
length of a randomly selected hair is

(i) [1 mark for the correct value; 2 marks for the work; total 3 marks] within 1.5 standard deviations from
its mean?

(ii) [1 mark for the correct value; 2 marks for the work; total 3 marks] farther than 2.5 standard deviations
from its mean?
Solution:
Let σ be the standard deviation of X, and let µ be the mean of X.
(i) If X is 1.5 standard deviations from its means, this implies µ − 1.5σ ≤ X ≤ µ + 1.5σ. It follows
that

P (µ − 1.5σ ≤ X ≤ µ + 1.5σ) = P (−1.5 ≤ Z ≤ 1.5)

= P (Z ≤ 1.5) − P (Z ≤ −1.5)
= Φ(1.5) − Φ(−1.5)
= 0.9332 − 0.0668
= 0.8664.

(ii) If X is farther than 2.5 standard deviations from its mean, this implies X > µ + 2.5σ OR
X < µ − 2.5σ. It follows that

P (X > µ + 2.5σ OR X < µ − 2.5σ) = P (X > µ + 2.5σ) + P (X < µ − 2.5σ)

= 1 − P (X ≤ µ + 2.5σ) + P (Z < −2.5)
= 1 − P (Z ≤ 2.5) + P (Z < −2.5)
= 1 − 0.9938 + 0.0062
= 0.0124.

Alternatively, the complement rule can be used; that is,

P (X > µ + 2.5σ OR X < µ − 2.5σ) = 1 − P (µ − 2.5σ ≤ X ≤ µ + 2.5σ)

5
 6. Suppose 10% of all ice cream cones made by a creamery is misshapen. Consider a random sample of
200 cones, and let X denote the number among these that are misshapen. Assume X is a binomial
random variable, and that it takes on the values x = 0, 1, 2, 3 . . . , 200.

(i) [total 2 marks] Explain why X can be assumed to have an approximately normal distribution.

(ii) [total 2 marks] Show the mean and variance of X is µ = 20 and σ 2 = 18, respectively.

(iii) [1 mark for the correct value; 2 marks for the work; total 3 marks] Use (ii) to determine the (approx-
imate) probability that X is at most 30?

(iv) [1 mark for the correct value; 2 marks for the work; total 3 marks] Use (ii) to determine the (approx-
imate) probability that X is less than 30?
Hint: since X takes on only integer values, then P (X < 30) = P (X ≤ 29).
Solution:
Let n = 200 and p = 0.10.
(i) Since np = 20 ≥ 10 and n(1 − p) = 200 ∗ 0.90 = 180 ≥ 10, then it we can assume X has an
approximately normal distribution.
(ii) The mean of X is µ = np = 20, and its variance is σ = np(1 − p) = 18.
(iii) We want to know P (X ≤ 30). Using a continuity correction, we have

P (X ≤ 30) ≈ P (X ≤ 30.5)
 
X −µ 30.5 − 20
=P ≤ √
σ 18
 
30.5 − 20
=P Z≤ √
18
= P (Z ≤ 2.47)
= Φ(2.47)
= 0.9932 from the z-table.

(iv) We want to know P (X < 30) = P (x ≤ 29). Using a continuity correction, we have

P (X < 30) = P (x ≤ 29) ≈ P (X ≤ 29.5)

 
X −µ 29.5 − 20
=P ≤ √
σ 18
 
29.5 − 20
=P Z≤ √
18
= P (Z ≤ 2.24)
= Φ(2.24)
= 0.9875 from the z-table.

7. The magnitude of earthquakes recorded in a region is modelled by an exponential distribution with

mean 2.4.

(i) [1 mark for the correct value; 2 marks for the work; total 3 marks] Find the probability an earthquake
striking this region will exceed 3.0.

6
(ii) [1 mark for the correct value; 2 marks for the work; total 3 marks] Find the probability an earthquake
striking this region will fall between 2.0 and 3.0.
Solution:
1
Let X=magnitude of earthquakes. Since µ = 2.4, then the parameter associated to X is λ = 2.4
.
(i) Since X follows an exponential distribution, the probability an earthquake striking this region will
exceed 3.0 is

P (X > 3.0) = 1 − P (X ≤ 3.0)

= 1 − F (3, λ)
= 1 − (1 − e−3λ )
= e−3λ
3
= e− 2.4 .

Alternatively, the probability can be determined using

Z ∞
P (X > 3.0) = f (x; λ)dx
3.0

where f is the probability density function of X.

(ii) Since X follows an exponential distribution, the probability an earthquake striking this region will
fall between 2.0 and 3.0

P (2.0 < X < 3.0) = P (X ≤ 3.0) − P (X < 2.0)

= F (3, λ) − F (2, λ)
= (1 − e−3λ ) − (1 − e−2λ )
= e−2λ − e−3λ
2 3
= e− 2.4 − e− 2.4 .

Note: P (2.0 ≤ X ≤ 3.0) is also acceptable, and will result in the same answer.
Alternatively, the probability can be determined using
Z 3.0
P (2.0 < X < 3.0) = f (x; λ)dx
2.0

where f is the probability density function of X.

8. Suppose the sediment density (g/cm) of a randomly selected specimen from a certain region is normally
distributed with mean µ = 2.65 and standard deviation σ = 0.85.
(a) [1 mark for the correct value; 2 marks for the work; total 3 marks] If a random sample of 25
specimens is selected, what is the probability that the sample mean sediment density is at most 3.00?
(b) [1 mark for the correct value; 2 marks for the work; total 3 marks] How large a sample size would
be required to ensure that the probability in part (a) is at least 0.99?
Solution:

7
 √
(a) We have n = 25, µX̄ = 2.65, and σX̄ = 0.85/ 25 = 0.17 and X̄ ∼ N (2.65, 0.17). Now we compute
 
X̄ − 2.65 3 − 2.65
P (X̄ ≤ 3) = P ≤ = P (Z ≤ 2.06) = 0.9803.
0.17 0.17

(b) We are asked to find n such that P (X̄ ≤ 3) ≥ 0.99. It follows that
   
X̄ − 2.65 3 − 2.65 3 − 2.65
P √ ≤ √ ≥ 0.99 =⇒ P Z ≤ √ ≥ 0.99.
0.85/ n 0.85/ n 0.85/ n

From the z-table,

3 − 2.65
√ ≥ 2.33
0.85/ n
2.33·0.85 2


from which n ≥ 0.35
= 32.02. Thus, n = 33 will suffice (n larger than 33 would also work).