University of Sydney

ECMT1010 Tutorial Solutions (2020, Semester 1, Week 13)

Lecturer: Tim Fisher

1. X = {1, 3, 5, 7, 8, 9}, Y = {2, 4, 7, 9} and Z = {1, 2, 3, 4, 7}

a. X or Z = {1, 2, 3, 4, 5, 7, 8, 9}
b. X and Y = {7, 9}
c. X and Z = {1, 3, 7}
d. X or Y or Z = {1, 2, 3, 4, 5, 7, 8, 9}
e. X and Y and Z = {7}
f. (X or Y) and Z = {1, 2, 3, 4, 5, 7, 8, 9} and {1, 2, 3, 4, 7} = {1, 2, 3, 4, 7}
g. (Y and Z) or (X and Y) = {2, 4, 7} or {7, 9} = {2, 4, 7, 9}
h. X or Y = {1, 2, 3, 4, 5, 7, 8, 9}
i. Y and X = {7, 9}

2. a. P(not A) = 1 – P(A) = 1 – 0.47 = 0.53

b. P(not B) = 1 – P(B) = 1 – 0.33 = 0.67
c. P(A or B) = P(A) + P(B) – P(A and B) = 0.47 + 0.33 – 0.20 = 0.60
d. P(A|B) = P(A and B) / P(B) = 0.20/0.35 = 0.606
e. P(B|A) = P(A and B) / P(A) = 0.20/0.47 = 0.426
f. A and B are not disjoint, because P(A and B) ≠ 0.

3. a. 18/121=0.149
b. 66/121=0.545
c. 11/121=0.09
d. 1-0.149=0.851
e. (66+38-21)/121=0.686
f. 24/66=0.364
g. 10/21=0.476
h. Yes. 20/44=55/121
i. Yes.
j. No.
4. Let C be the event that a person has the condition, so (not C) is the event that the person
does not have the condition. Let P be the event of testing positive for the condition. Then
by Bayes’ rule:
P (C ) × P ( P | C )
P (C | P ) =
P ( C ) × P ( P | C ) + P ( not C ) × P ( P | not C )
0.0001× 0.99
= = 0.0098.
0.0001× 0.99 + 0.9999 × 0.01

So even if someone tests positive for the condition, there is a less than 1% chance that
they actually have it. Thus, if you screen a large number of people you will get a vast
number of false positives. Screening programs therefore should only target parts of the
population that are genuinely at risk of having the condition in question.

5. For this question the following information is relevant:

S = Scarlet is the murderer
M = Mustard is the murderer
P = Plum is the murderer
K = Knife is the murder weapon
L = Lead pipe is the murder weapon
R = Rope is the murder weapon

1
𝑃𝑃(𝑆𝑆) = 𝑃𝑃(𝑀𝑀) = 𝑃𝑃(𝑃𝑃) =
3
𝑃𝑃(𝐾𝐾|𝑆𝑆) = 0.4 𝑃𝑃(𝐾𝐾|𝑀𝑀) = 0.2 𝑃𝑃(𝐾𝐾|𝑃𝑃) = 0.3
𝑃𝑃(𝐿𝐿|𝑆𝑆) = 0.35 𝑃𝑃(𝐿𝐿|𝑀𝑀) = 0.3 𝑃𝑃(𝐿𝐿|𝑃𝑃) = 0.4
𝑃𝑃(𝑅𝑅|𝑆𝑆) = 0.25 𝑃𝑃(𝑅𝑅|𝑀𝑀) = 0.5 𝑃𝑃(𝑅𝑅|𝑃𝑃) = 0.3

1
a. 𝑃𝑃(𝐾𝐾) = 𝑃𝑃(𝐾𝐾|𝑆𝑆)𝑃𝑃(𝑆𝑆) + 𝑃𝑃(𝐾𝐾|𝑀𝑀)𝑃𝑃(𝑀𝑀) + 𝑃𝑃(𝐾𝐾|𝑃𝑃)𝑃𝑃(𝑃𝑃) = (0.4 + 0.2 + 0.3) = 0.30
3
1
b. 𝑃𝑃(𝐿𝐿) = 𝑃𝑃(𝐿𝐿|𝑆𝑆)𝑃𝑃(𝑆𝑆) + 𝑃𝑃(𝐿𝐿|𝑀𝑀)𝑃𝑃(𝑀𝑀) + 𝑃𝑃(𝐿𝐿|𝑃𝑃)𝑃𝑃(𝑃𝑃) = (0.35 + 0.3 + 0.4) = 0.35
3
1
c. 𝑃𝑃(𝑅𝑅) = 𝑃𝑃(𝑅𝑅|𝑆𝑆)𝑃𝑃(𝑆𝑆) + 𝑃𝑃(𝑅𝑅|𝑀𝑀)𝑃𝑃(𝑀𝑀) + 𝑃𝑃(𝑅𝑅|𝑃𝑃)𝑃𝑃(𝑃𝑃) = (0.25 + 0.5 + 0.3) = 0.35
3
𝑃𝑃(𝑀𝑀∩𝐿𝐿) 𝑃𝑃�𝐿𝐿�𝑀𝑀 �𝑃𝑃(𝑀𝑀) (0.3/3)
d. 𝑃𝑃(𝑀𝑀|𝐿𝐿) = = = = 0.286
𝑃𝑃(𝐿𝐿) 𝑃𝑃(𝐿𝐿) 0.35
𝑃𝑃(𝑆𝑆∩𝑅𝑅) 𝑃𝑃�𝑅𝑅 �𝑆𝑆�𝑃𝑃(𝑆𝑆) (0.25/3)
e. 𝑃𝑃(𝑆𝑆|𝑅𝑅) = = = = 0.238
𝑃𝑃(𝑅𝑅) 𝑃𝑃(𝑅𝑅) 0.35
𝑃𝑃(𝑀𝑀∩𝑅𝑅) 𝑃𝑃�𝑅𝑅 �𝑀𝑀�𝑃𝑃(𝑀𝑀) (0.5/3)
f. 𝑃𝑃(𝑀𝑀|𝑅𝑅) = = = = 0.476
𝑃𝑃(𝑅𝑅) 𝑃𝑃(𝑅𝑅) 0.35
𝑃𝑃(𝑃𝑃∩𝐾𝐾) 𝑃𝑃�𝐾𝐾 �𝑃𝑃�𝑃𝑃(𝑃𝑃) (0.3/3)
g. 𝑃𝑃(𝑃𝑃|𝐾𝐾) = = = = 0.333
𝑃𝑃(𝐾𝐾) 𝑃𝑃(𝐾𝐾) 0.30
𝑃𝑃(𝑆𝑆∩𝐾𝐾) 𝑃𝑃�𝐾𝐾 �𝑆𝑆 �𝑃𝑃(𝑆𝑆) (0.4/3)
h. 𝑃𝑃(𝑆𝑆|𝐾𝐾) = = = = 0.444
𝑃𝑃(𝐾𝐾) 𝑃𝑃(𝐾𝐾) 0.30
𝑃𝑃(𝑃𝑃∩𝐿𝐿) 𝑃𝑃�𝐿𝐿�𝑃𝑃�𝑃𝑃(𝑃𝑃) (0.4/3)
i. 𝑃𝑃(𝑃𝑃|𝐿𝐿) = = = = 0.381
𝑃𝑃(𝐿𝐿) 𝑃𝑃(𝐿𝐿) 0.35
6. Let M = Mastercard, A = American Express, V = Visa.
Given: P(M) = .30, P(A) = .20, P(V) = .25,
P(M and A) = .08, P(V and M) = .12, P(A and V) = .06.

a. P(V or A) = P(V) + P(A) - P(V and A) =.25 + .20 - .06 = .39.

b. P(V | M) = P(V and M) / P(M) = .12/.30 = .40.
c. P(M | V) = P(V and M) / P(V) = .12/.25 = .48.
d. Is P(V) = P(V | M)?
Check: .25 ≠ .40 .
So, Possession of Visa is not independent of possession of Mastercard.
e. American Express is not mutually exclusive of Visa because P(A and V) ≠ 0.

7. a. By additive rule, we obtain the following two equations.

(1)…………… 0.75 = P(A or B) = P(A) + P(B) – P(A and B);
(2)…………… 0.80 = P(A or (not B)) = P(A) + P(not B) – P(A and (not B)).

Adding the two equations together, we have

(1) + (2)………. 1.55 = 2P(A) + [ P(B) + P(not B) ] – [ P(A and B) + P(A and (not B)) ].

By the law of total probabilities, it follows that

1.55 = 2P(A) + 1 – P(A),
P(A) = 0.55.

b. That A and B are mutually exclusive implies P(A or B) = P(A) + P(B).

As a result, we have
0.75 = 0.55 + P(B),
P(B) = 0.2.

c. That A and B are independent implies P(A and B) = P(A)P(B).

Also, by (1), 0.75 = P(A) + P(B) – P(A and B).
Therefore, we have 0.75 = P(A) + P(B) – P(A)P(B).
Substituting part (a)’s result, we obtain
0.75 = 0.55 + P(B) – 0.55P(B),
P(B) = 0.2/0.45 = 4/9.
8. a. p(1) + p(2) = 0.507

b. p(0) + p(1) + p(2) = 0.990

c. 1- p(0) - p(1) = 0.114

d. The mean number of correct picks on a “Take Five” ticket is

5
µ= ∑ xp( x) = 0 × 0.483287 + 1× 0.402739 + 2 × 0.103933 + 3 × 0.009744
x =0

+4 × 0.000295 + 5 × 0.000002 ≈ 0.641

e. The variance of the number of correct picks on a "Take Five" ticket is

5

∑ ( x − µ ) p( x) =
σ2 = ( 0 − 0.641) × 0.483287 + (1 − 0.641) × 0.402739
2 2 2

x =0

+ ( 2 − 0.641) × 0.103933 + ( 3 − 0.641) × 0.009744

2 2

+ ( 4 − 0.641) × 0.000295 + ( 5 − 0.641) × 0.000002 ≈ 0.500

2 2

f. Compute the standard deviation of the number of correct picks on a "Take Five" ticket.
σ
= σ2
= 0.50 ≈ 0.707

9. a. X is a binomial random variable because

i) the number of trials is fixed in advanced (n = 10);
ii) the probability of success (obese) is the same for all trials (p = 0.254);
iii) since this is a random sample, the trials (individuals selected for the sample) are
supposed to be independent.

b. The probability that exactly 7 people are obese is

10 
  ( 0.254 ) (1 − 0.254 ) 120 × ( 0.254 ) ( 0.746 ) ≈ 0.0034
10 − 7
P( X =
7) = =
7 7 3

7
c. The probability that more than three people are obese is
P( X > 3) =
1 − P ( X ≤ 3) =
1 − P( X =
0) − P ( X =
1) − P ( X =
2) − P ( X =
3)
10  10  10 
= 1 −   ( 0.254 ) ( 0.746 ) −   ( 0.254 ) ( 0.746 ) −   ( 0.254 ) ( 0.746 )
0 10 1 9 2 8

0 1 2

10 
−   ( 0.254 ) ( 0.746 )
3 7

3
≈ 0.23

=
d. The mean number of obese residents in the sample is µ 10
= ( 0.254 ) 2.54
is σ
e. The standard deviation= 10 ( 0.254 )(1 − 0.254 ) ≈ 1.89
10. a. All of the ten digits are equally likely to be drawn by the computer, with a probability of
0.1 each. Therefore, the probability of “success” is P(success) = P(the random value is 5,
6, 7, 8 or 9) = 0.5.

b. There are 4 independent trials, each with a probability of success 0.5. Let X = the number
of random values that are 5 or larger. We can thus model X using a binomial probability
distribution with n=5 and p = P(success) = 0.5. In short, X  Bin= (n 4,= p 0.5) .

The probability that all 4 values are 5 or large is

 4
  ( 0.5 ) (1 − 0.5 ) =
4− 4
P( X =
4) = 0.54 ≈ 0.063
4

 
4

c. The probability that 2 or more of the values are 5 or larger is

P( X =
2,3 or 4) =
1 − P( X =
0 or 1)
 4  4
1 −   ( 0.5 ) (1 − 0.5 ) −   ( 0.5 ) (1 − 0.5 )
4−0 4 −1
=
0 1

0 1
1 − 0.54 − 4 ( 0.5 ) ≈ 0.688
=
4

d. The probability that 2 or 3 of the values are 5 or larger is

P( X = 2 or 3)
 4  4
=   ( 0.5 ) (1 − 0.5 ) +   ( 0.5 ) (1 − 0.5 )
2 4− 2 3 4 −3

 2  3
= 6 ( 0.5 ) + 4 ( 0.5 ) ≈ 0.625
4 4

= 4 ( 0.5=
µ np
e. The mean of X is = ) 2.

f. The standard deviation of X is σ

= np (1 − p=
) 4 ( 0.5 )(1 − 0.5=
) 1.

11. a. Let µ a be the mean number of passengers arriving at the airport terminal for a small jet
flight.

Since the total number of flight tickets sold is fixed at 32 and the ticket holders will show
up independently with probability 0.1 each, we can model the number of passengers
arriving the airport terminal, X, with a binomial distribution, with n = 32 and p = 0.9, so
the mean would be µ a = np = 32(0.9) = 28.8. Let p=
( x) P=
( X x) be the binomial
probability mass function for X.

b. Everyone gets a seat when X ≤ 30. To find this probability we use the complement rule
(find the chance too many people show up with X = 31 or X = 32, then subtract from
one.)
 P( X ≤ 30)
=1 − P( X =−
31) P( X =
32)
 32   32 
= 1 −   ( 0.9 ) ( 0.1) −   ( 0.9 ) ( 0.1)
31 32 0

 31   32 
≈ 1 − 0.122 − 0.034
≈ 0.844

Everyone will have a seat on about 84.4% of the flights. The airline will need to deal with
overbooked passengers on the other 15.6% of the flights.

c. Let µb be the mean number of passengers on board for a small jet flight.
30
From the first principle, µb = ∑ xp( x) + 30 p(31) + 32 p(32) .
x =1
Since each small jet has

30 seats, in order to get µb we only need to adjust the last two terms of µ a , as follows.
30
µb = ∑ xp( x) + 30 p(31) + 30 p(32)
x =1
32
= ∑ xp( x) − 31 p(31) − 32 p(32) + 30 p(31) + 30 p(32)
x =1

µa − p(31) − 2 p(32)
=
 32 
≈ 28.8 −   ( 0.9 ) ( 0.1) − 2 ( 0.9 )
31 32

 31 
≈ 28.8 − 0.122 − 2 ( 0.034 ) ≈ 28.6