Assignment 1 Probability

Due: 23:59 September 22 (Sunday), via Moodle

Problem 1 (Market Researchers Inc.)

Market Researchers Inc. has been hired to perform a study to determine if the market for a new
product will be good or poor. In similar studies conducted in the past, whenever the market
actually was good, the market research study indicated that it would be good 80% of the time.
On the other hand, whenever the market actually was poor, the market study incorrectly
predicted it would be good 20% of the time. Before the study is performed, it is believed there is
a 30% chance the market will be good. When Market Researchers, Inc. performs the study for
this product, the results predict the market will be good. Given the results of this study, what is
the probability that the market actually will be good?

Solution:

Events Optimistic Prediction (O) Negative Prediction Prior belief

(N)

Good Market (G) P(G)*P(O|G) P(G) = 0.3

=0.3*0.8=0.24

Poor Market (P) P(P)*P(O|P) P(P) = 0.7

=0.7*0.2=0.14

Therefore, P(G|O) = P(G and O)/P(O) = 0.24/(0.24+0.14) = 0.63

(Note the other two joint probabilities, P(G&N) and P(P&N), are not necessary for the answer;
nevertheless, they can be deduced from the information given.)
Problem 2. The Springfield Kings
The Springfield Kings, a professional basketball team, has won 15 of its last 20 games and is
expected to continue winning at the same percentage rate. The team’s ticket manager is
anxious to attract a large crowd to tomorrow’s game but believes that depends on how well the
Kings perform tonight against the Galveston Comets. He assesses the probability of drawing a
large crowd to be 0.8 should the team win tonight. What is the probability that the team wins
tonight and that there will be a large crowd at tomorrow’s game?

Solution:

The probability of winning tonight’s game is

number of wins /0
number of games
= 12 = 0.75

The probability that the team wins tonight is 0.75. The probability that the team wins tonight and
draws a large crowd at tomorrow’s game is a joint probability of dependent events. Let the
probability of winning be P(W) and the probability of drawing a large crowd be P(L). Thus
P(WL) = P(L | W) ´ P(W) (the probability of large
crowd is 0.8 if the team
= 0.8 ´ 0.75
wins tonight)
= 0.60
Thus, the probability of the team winning tonight and of there being a large crowd at tomorrow’s
game is 0.60.
Problem 3 (Lost Soldier)
The oasis outpost of Abu Ilan, in the heart of the Negev desert, has a population of 40 Bedouin
tribesmen and 60 Farima tribesmen. El Kamin, a nearby oasis, has a population of 30 Bedouins
and 30 Farima. A lost Israeli soldier, accidentally separated from his army unit, is wandering
through the desert and arrives at the edge of one of the oases. The soldier has no idea which
oasis he has found, but the first person he spots at a distance is a Bedouin. What is the probability
that he wandered into Abu Ilan? What is the probability that he is in El Kamin?

Solution:

The probability that the soldier is in either Abu Ilan or El Kamin is 0.5.

Events Spotting a Bedouin (B) Spotting a Farima (F) Prior belief

Being in Abu Ilan (A) P(A)*P(B|A) P(A)*P(F|A) P(A) = 0.5

=0.5*40/(40+60)=0.2 =0.5*60/(40+60)=0.3

Being in El Kamin (E) P(E)*P(B|E) P(E)*P(F|E) P(E) = 0.5

=0.5*30/(30+30)=0.25 =0.5*30/(30+30)=0.25

P(B) = 0.2+0.25 = 0.45 P(F) = 0.3+0.25 = 0.55

As a result, P(A|B)=0.2/0.45= 0.4444 and P(E|B)= 0.25/0.45=0.5556.

Problem 4. (A Construction Project)
The time to complete a construction project is normally distributed with a mean of 52 weeks and
a standard deviation of 5 weeks. (a) What is the probability the project will be finished in 60
weeks or less? (b) What is the probability the project will be finished in 66 weeks or less? (c) What
is the probability the project will take longer than 55 weeks?

Solution:

Let X denote the weeks it takes to finish the project. We treat X as a continuous random variable.

(1) P(X<=60) = P((X-52)/5<=8/5) = P(Z<=8/5) = 0.9452

(2) P(X<=66) = P((X-52)/5<=14/5) = P(Z<=14/5) = 0.9974
(3) P(X>55) = P((X-60)/5> 3/5) = P(Z>3/5) = 1 - P(Z<=3/5) = 1 – 0.7257 = 0.2743
Problem 5. (Lisa's Hair Salon)
Customers arrive at Lisa’s Hair Salon during a day at any particular business hour according to a
Poisson distribution with a rate of 𝜆 = 3 per hour. Now, the salon has two barbers who can each
service a customer in exactly 30 minutes. Suppose a customer, Amy, arrives at 2:00 pm and finds
both barbers idle.
(a) What is the probability that we will observe customers waiting before 2:30 pm?
(b) What is the probability that Amy will observe the next customer’s (i.e., the customer after
her) arrival before 2:15 pm?
(c) What is the probability that Amy will find the shop empty when she leaves?

Solution:

(a) The event that customers have to wait before 2:30pm is equivalent to the event that at least two
more customers arrive after 2:00pm and before 2:30pm. This is because the customer arriving at
2:00pm occupies one barber until 2:30pm and the next customer will occupy the other barber.
Customers arriving thereafter and before 2:30pm will have to wait.
The event above is equivalent to the event “at least two customers come between 2:00pm and
2:30pm” (recall, we do not count the arrival (i.e. Amy) on 2:00pm), which is further translated to
“𝑋 ≥ 2”, where X follows Poisson distribution with l’ = 1.5 (since we are looking at a 30-min
time window).

𝑃(𝑋 ≥ 2) = 1 − 𝑃(𝑋 ≤ 1) = 1 − 0.5578 = 𝟎. 𝟒𝟒𝟐𝟐

(b) Two ways to do this.

Method 1: the event is equivalent to “we are looking at a 15-min time window and observe at
least one arrival” = “𝑋 ≥ 1”, where X follows Poisson(0.75).
Method 2: the event is equivalent to “the inter-arrival time between Amy and the next customer
is less than 15 min” = “𝑌 ≤ 0.25”, where Y follows Exponential (3).
Then the answer is P(𝑋 ≥ 1) or P(𝑌 ≤ 0.25), both taking value 0.5276.

(c) This is equivalent to the event that no customers arriving in 30 minutes. Thus, for X following
Poisson(1.5),
𝑃(𝑋 = 0) = 𝑒 !".$ = 𝟎. 𝟐𝟐𝟑𝟏
Problem 6. (Armstrong Faber)
Armstrong Faber produces a standard number-two pencil called Ultra-Lite. The demand
for Ultra-Lite has been stable over the past 6 years. On average, Armstrong Faber
has sold 957,000 pencils each year. Furthermore, 90% of the time, sales have been between
954,000 and 960,000 pencils. The sales are expected to follow a normal distribution with a mean
of 957,000 pencils. Estimate the standard deviation of this distribution. (Hint: you may need the
Standard Normal Table attached to this assignment sheet.)

Solution:

Let X denote the annual sales volume. (Here we treat X as a continuous random variable.) Let s denote
the standard deviation.

We know

0.9 = P(954,000<X<960,000)

= P(X<960,000) – P(X<954,000)

= P(Z<3,000/s) – P(Z<-3,000/s)

= P(Z<3,000/s) – P(Z>3,000/s)

= P(Z<3,000/s) – [1 – P(Z<3,000/s)]

= 2*P(Z<3,000/s) – 1

Therefore, we have P(Z<3,000/s)=1.9/2=0.95; thus, 3,000/s = 1.645 and s = 1,823.7.