N ATIONAL U NIVERSITY OF S INGAPORE

D EPARTMENT OF S TATISTICS AND DATA S CIENCE

ST2334 P ROBABILITY AND S TATISTICS
F INAL E XAMINATION S AMPLE PAPER 1
(S EMESTER II, AY 2024/2025)
T IME A LLOWED : 120 MINUTES

Suggested Solutions

I NSTRUCTIONS TO S TUDENTS

1. Please write your student number only. Do not write your name.

2. This assessment contains 30 questions and comprises 1 printed pages.

3. The total marks is 60; marks are equal distributed for all questions.

4. Please answer ALL questions.

5. Calculators may be used.

6. This is an OPEN BOOK assessment. Only HARD COPIES of materials are allowed.
 ST2334

1. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY

A random variable X has the following probability function.
x
fX (x) = , for x = 0.4, 0.6, 0.9, 2.1;
4
and fX (x) = 0 elsewhere.
What type of random variable is X?

(a) Continuous (c) Unable to determine

(b) Discrete (d) None of the given options

S OLUTION
(b)
2. T RUE /FALSE
Suppose X ∼ N(1, σ12 ) and Y ∼ N(0, σ22 ). P(X < 1) is larger than P(Y < 1) when σ12 >
σ22 > 0.
• TRUE
• FALSE
S OLUTION
False
No matter what are the values for σ1 and σ2 , we must have P(X < 1) = 0.5 and P(Y <
1) > 0.5.
3. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY
A professor receives, on average, 21.7 emails from students the day before the midterm
exam. To compute the probability of receiving at least 10 emails on such day, what type
of probability distribution will he use?

(a) Binomial distribution. (c) Normal distribution.

(b) Poisson distribution. (d) Negative Binomial distribution.

S OLUTION
(b)
4. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY
Which of the following would be an appropriate null hypothesis?

(a) The mean of a population is equal to 55.

(b) The mean of a sample is equal to 55.
(c) The mean of a population is greater than 55.
(d) None of the given options

S OLUTION
(a)
5. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY
Let A, B be events in sample space S. Which of the following may NOT be true?

2
 ST2334

(a) A ∩ A′ = 0/ (c) (A ∪ B)′ = A′ ∪ B′

(b) A ∪ A′ = S. (d) A ∪ B = A ∪ (B ∩ A′ )

S OLUTION
(c)

6. F ILL IN THE BLANK

Let X and Y be independent random variables such that E(X) = 1, E(Y ) = 2, V (X) = 3,
V (Y ) = 4. Compute V (2X −Y ).
Answer:
(Provide your answer in decimal form and round it to two decimal places if necessary.)
S OLUTION
16

7. F ILL IN THE BLANK

We toss a fair die until the outcome “6” appears twice. Find the probability that it takes
5 tosses.
Answer:
(Provide your answer in decimal form and round it to three decimal places if necessary.)
S OLUTION
     2  3
1 4 1 5
X ∼ NB 2, . P(X = 5) = = 0.064.
6 1 6 6
8. T RUE /FALSE
Under the usual random sampling setup, we can halve the standard deviation of the sam-
ple mean by doubling the sample size.

• TRUE
• FALSE

S OLUTION
False

9. T RUE /FALSE
Let (X,Y ) be a random vector, then for any real numbers x and y, we must have

P(X ≤ x,Y ≤ y) = 1 − P(X > x,Y > y).

• TRUE
• FALSE

S OLUTION
FALSE.

10. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY

Which of the following is a valid cumulative distribution function?

3
 ST2334

 

 0 x ≤ −1 
 0 x < −1
0.3 −1 < x ≤ 1 0.6 1 < x ≤ 10
 
(a) F(x) = (c) F(x) =

 0.7 1 < x ≤ 10 
 0.7 −1 < x ≤ 1
1 elsewhere 1 elsewhere
 
 

 0 x < −1 
 0 x < −1
0.5 −1 < x < 1 0.6 −1 ≤ x < 1
 
(b) F(x) = (d) F(x) =

 0.7 1 < x < 10 
 0.7 1 ≤ x < 10
1 elsewhere 1 elsewhere
 

S OLUTION
(d)

11. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY

The life time (in years) of a certain brand of light bulb follows an exponential distribution
1
with the probability density function: exp(−x/2). What is the probability that the bulb
2
will last for more than 5 years, given that it has been working for 3 years?
1 1
(a) exp(−5/2) (c) exp(−1)
2 2
(b) exp(−5/2) (d) exp(−1)

S OLUTION
(d)

12. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY

The Central Limit Theorem is important in statistics because

(a) for a large sample size, n, it says the population is approximately normal.
(b) for any population, it says the sampling distribution of the sample mean is approxi-
mately normal, regardless of the sample size.
(c) for a large sample size, n, it says the sampling distribution of the sample mean is
approximately normal, regardless of the shape of the population.
(d) for any sized sample, it says the sampling distribution of the sample mean is ap-
proximately normal.

S OLUTION
(c)

13. F ILL IN THE BLANK

John rolls a fair die 6 times independently. What is the probability that he will get num-
bers more than 2 at least twice?

Answer:
(Provide your answer in decimal form and round it to three decimal places if necessary.)
S OLUTION
Let X = number of times to get numbers more than 2. X ∼ Bin(6, 2/3).
P(X ≥ 2) = 1 − (P(X = 0) + P(X = 1)) = 1 − (0.001372 + 0.016461) = 0.9822.

4
 ST2334

14. F ILL IN THE BLANK

Jill’s bowling scores are approximately normally distributed with mean 170 and standard
deviation 20, while Jack’s bowling scores are approximately normally distributed with
mean 160 and standard deviation 15. If Jack and Jill each bowl one game, then assuming
that their scores are independent random variables, the probability that the sum of the
scores is higher than 340 is approximately equal to Φ(c). Find the value of c.

Answer: c =
(Provide your answer in decimal form and round it to three decimal places if necessary.)
Note: Φ(·) denotes the cumulative distribution function of N(0, 1).
S OLUTION
−0.4
Denote Y = X1 + X2 ∼ N(170 + 160, 202 + 152 ) = N(330, 252 ).
 
Y − 330
P(Y > 340) = P > 0.4 = P(Z < −0.4).
25

15. F ILL IN THE BLANK

An experiment was carried out to test whether mean weight gain for pigs fed ration A is higher
than those fed ration B. Eight pairs of pigs were used. The rations were assigned at random
to the two animals within each pair. The gain (in kilograms) after 45 days, assuming normally
distributed, are given as follows.

Pairs 1 2 3 4 5 6 7 8 mean sd
Ration A 30 17 18 21 22 30 24 27 23.625 5.0409
Ration B 26 18 15 20 21 25 27 23 21.875 4.1555
Difference, A - B 4 -1 3 1 1 5 -3 4 1.75 2.7646

Suppose that the pigs within each pair were littermates. What is the observed value of the test
statistic in testing the alternative hypothesis that ration A is better, in terms of mean weight gain,
than ration B at a 5% significance level?
Answer:
(Provide your answer in decimal form and round it to two decimal places if necessary.)
S OLUTION
D √
We have the paired data. T = √ = 1.75/(2.7646/ 8) = 1.7904.
SD / n
16. F ILL IN THE BLANK
The mean lifetime of 100 randomly selected pumps made by a particular factory was 200
days. Assuming it is known that the population standard deviation σ = 40, find a 95%
confidence interval for the mean lifetime of pumps made by the factory.
Answer: ( , ).
Note: z0.025 = 1.96; z0.05 = 1.64.
(Provide your answers in decimal form and round them to two decimal places if neces-
sary.)

5
 ST2334

S OLUTION
x̄ = 200; n = 100 is large; σ = 40 is known. Thus the 95% confidence interval is
σ 40
x̄ ± z0.025 √ = 200 ± 1.96 √ = (192.16, 207.84).
n 100

17. F ILL IN THE BLANK

We roll a fair die 3 times. Find the probability that the sum is equal to 5.
Answer:
(Provide your answer in decimal form and round it to three decimal places if necessary.)
S OLUTION
The possible three numbers are {1, 1, 3} or {1, 2, 2}; this leads to six possibilities:

(1, 1, 3), (1, 3, 1), (3, 1, 1); (1, 2, 2), (2, 1, 2), (2, 2, 1).

In total there are 63 = 216 possibilities. Therefore the probability is 6/216 = 1/36 =
0.0278.

18. T RUE /FALSE

Consider the Z-test for H0 : µ = 0 based on X1 , . . . , Xn i.i.d. N(µ, σ 2 ). It turns out that
X̄ = 2.3. The p-value for the one-sided test (H1 : µ > 0) is half of that for the two-sided
test (H1 : µ ̸= 0).

• TRUE
• FALSE

S OLUTION
True

19. F ILL IN THE BLANK

A new COVID rapid test is able to correctly diagnose that you do not have the virus 90%
of the time. However, if you do have the virus, it fails to detect it 25% of the time. Given
that the overall COVID infection rate at a particular worker dorm is 20%, what is the
probability of a worker being infected if his rapid test does not detect the virus?
Answer:
(Provide your answer in decimal form and round it to three decimal places if necessary.)
S OLUTION

P(T ′ |D′ ) = 0.9, P(T ′ |D) = 0.25, P(D) = 0.2

P(T ′ |D)P(D) (0.25)(0.2)
P(D|T ′ ) = ′ ′ ′ ′
= = 0.0649.
P(T |D)P(D) + P(T |D )P(D ) (0.25)(0.2) + (0.9)(0.8)

20. M ULTIPLE R ESPONSE : C HOOSE ALL ANSWERS THAT APPLY

Which of the following can happen if the hypothesis is rejected.

(a) p-value> α;

6
 ST2334

(b) test statistic falls in the rejection region;

(c) type I error occurs;
(d) type II error occurs.

S OLUTION
(b), (c).

21. F ILL IN THE BLANK

Let X1 , X2 , . . . , X100 be independent and identically distributed continuous random
! vari-
100
ables with E(Xi ) = 5 and V (Xi ) = 4. Compute approximately P ∑ Xi > 510 .
i=1
Answer:
(Provide your answer in decimal form and round it to three decimal places if necessary.)
Note: Φ(0.5) = 0.6915, Φ(1) = 0.8413, Φ(1.5) = 0.9332; where Φ(·) denotes the cu-
mulative distribution function for N(0, 1).
S OLUTION
0.3085
100
By CLT, ∑ Xi ≈ N(500, 400).
i=1 !
100  
510 − 500
P ∑ Xi > 510 =P Z> √ = P(Z > 0.5) = 1 − 0.6915 = 0.3085.
i=1 400

22. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY

Which of the following statements about probability is INCORRECT?

(a) If A and B are two events and P(A ∩ B) = P(A), then P(A ∩ B′ ) = 0.
(b) Let S be the sample space and let A be an event. If there exists an x ∈ S but x ∈
/ A,
then P(A) < 1.
(c) Let A and B be two events; then P(A ∪ B) ≤ P(A) + P(B).
(d) Let A and B be independent events. If P(A) > 0 and P(B) > 0, then A and B are not
mutually exclusive.

S OLUTION
(b)
23. F ILL IN THE BLANK
Consider the following game:
• First round: the gamer flips a fair coin. If he gets a head, he loses; otherwise he wins the
round.
• Second round: the gamer flips two fair coins independently. If he gets two heads, he loses;
otherwise he wins the round.
• Third round: the gamer flips three fair coins independently. If he gets three heads, he loses;
otherwise he wins the round.
• And so on.

7
 ST2334

What is the probability that the gamer will make his first win in the 4th round?
Note: you can assume that from rounds to rounds, the flips are independently conducted.
Answer:
(Provide your answer in decimal form and round it to four decimal places if necessary.)
S OLUTION
Let Ai = {the gamer wins the ith round}. The question is asking

1 1 1 15
P(A′1 ∩ A′2 ∩ A′3 ∩ A4 ) = P(A′1 )P(A′2 )P(A′3 )P(A4 ) = · · · = 0.01464844.
2 4 8 16

24. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY

Let {X1 , X2 , . . . , X25 } be a random sample from the N(µ, 22 ) distribution. Consider the
X̄
hypotheses: H0 : µ = 0 versus H1 : µ ̸= 0, and the test statistic Z = √ . Suppose that
2/ 25
we reject H0 if |zobs | > 2; and do not reject H0 otherwise. What is the probability that we
will not reject H0 given that the true value of µ is equal to 2?
Note: Φ(z) denotes the c.d.f. of the standard normal distribution: Φ(z) = Pr(Ze ≤ z) with
Ze ∼ N(0, 1).

(a) Φ(−3) − Φ(−7) (c) Φ(−2) − Φ(−6)

(b) Φ(2) − Φ(−2) (d) Φ(−3) − Φ(−8)

S OLUTION
(a)
We are to compute the type II error probability:
 
X̄
Pr(|Z| ≤ 2|µ = 2) = Pr −2 ≤ ≤2 µ =2
2/5
 
2 X̄ − 2 2
= Pr −2 − ≤ ≤ 2− µ =2
2/5 2/5 2/5
= Pr(−7 ≤ Ze ≤ −3) = Φ(−3) − Φ(−7).

25. T RUE /FALSE

For any θ ∈ R, the function

 θ − x if θ − 1 ≤ x < θ
f (x) = x − θ if θ ≤ x ≤ θ + 1
0 elsewhere


can serve as a probability density function of some distribution, whose population mean
is equal to θ .

• TRUE
• FALSE

8
 ST2334

S OLUTION
True

26. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY

In a course, students are graded based on a “normal curve”. For example, students within
0.5 standard deviation from the mean receive a C; between 0.5 and 1.0 standard deviation
above the mean receive a C+; between 1.0 and 1.5 standard deviation above the mean
receive a B; between 1.5 and 2.0 standard deviation receive a B+, etc. The class average
in an exam was 60 with a standard deviation of 10. What are the bounds for a B grade
and the percentage of students who will receive a B grade?

(a) (65, 75), 24.17% (c) (70, 75), 18.38%

(b) (65, 75), 12.08% (d) (70, 75), 9.19%

Note: Φ(1) = 0.8413, Φ(1.5) = 0.9332, Φ(2) = 0.9772, where Φ(·) denotes the cumu-
lative distribution of N(0, 1).
S OLUTION
(d)

27. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY

Flip an unfair coin. If a head shows, roll a fair die and report the number; otherwise, roll
a fair die twice and report the summation minus 1. Then P(6 is reported) =?

(a) 1/6 (c) 1/2

(b) 1/4 (d) can not tell

S OLUTION
(a)
Let X = 1 if coin shows head; Y = number obtained from rolling the die (once, or sum-
mation of twice).

P(6 is reported) = P({X = 1,Y = 6} or {X = 0,Y = 7})

= P(X = 1)P(Y = 6|X = 1) + P(X = 0)P(Y = 7|X = 0)
= p(1/6) + (1 − p)(1/6) = 1/6.

Note that

P(Y = 7|X = 0) = P(rolling two dice and get sum = 7)

= P((1, 6) or (2, 5) or (3, 4) or (4, 3) or (5, 2) or (6, 1))
= 6/36 = 1/6.

28. F ILL IN THE BLANK

An urn contains 3 red balls and 5 white balls. 4 balls are drawn uniformly at random
without replacement from the urn. Let X be the random variable for the number of red
balls drawn. If X ≤ 1, you win $X. If X > 1, you flip a fair coin. If the coin comes up
heads, you double your winnings and win a total of $2X. If the coin comes up tails, you

9
 ST2334

still win $X. Let W be the random variable for your winnings. Find the probability that
W is odd.
Answer:
(Provide your answer in decimal form and round it to four decimal places if necessary.)
S OLUTION
0.4642
X can take on values from {0, 1, 2, 3}, so W can only take on values from {0, 1, 2, 3, 4, 6}.
P(W is odd) = P(W = 1) + P(W = 3)
= P(X = 1) + P(X = 3 and coin flips tail)
3
5
3
5

1
= 1 8
3 + 3 8
1 ·
4 4
2
3 1 1
= + ·
7 14 2
13
= = 0.4642.
28

29. M ULTIPLE C HOICE : C HOOSE ONE ANSWER ONLY

Let X be a random variable with density function
 √
 k x , for 0 ≤ x ≤ 1
1−x
f (x) = ke 2 , for x > 1

0 , otherwise
where k is a constant. What is the value of k?

(a) 1/2 (c) 3/8

(b) 2/5 (d) 4/9

S OLUTION
(c)
Z ∞ Z 1
√ Z ∞
1−x
f (x) dx = k x dx + ke 2
−∞ 0 1
2 3 1 h
 
1−x ∞
i
= kx 2 + −2ke 2
3 0 1
2 8
= k − (−2k) = k
3 3
8 3
⇒ k=1⇒k= .
3 8

30. M ULTIPLE R ESPONSE : C HOOSE ALL ANSWERS THAT APPLY

Let X1 , X2 , . . . , Xn1 be independent and identically distributed (i.i.d.) random variables
with population mean µ1 ; let Y1 ,Y2 , . . . ,Yn2 be i.i.d. random variables with population
mean µ2 ; let U1 ,U2 , . . . ,Un3 be i.i.d. random variables with population mean 4. All these
random variables have the unknown but common variance σ 2 . Which of the following
is/are unbiased estimator(s) for σ 2 ?

10
 ST2334

n1
∑i=1 (Xi − X̄)2 + ∑nj=1
2
(Y j − Ȳ )2 + ∑nk=1
3
(Uk − 4)2
(a) .
n1 + n2 + n3 − 2
n1
∑i=1 (Xi − X̄)2 + ∑nj=1
2
(Y j − Ȳ )2 + ∑nk=1
3
(Uk − 4)2
(b) .
n1 + n2 + n3 − 3
n n n
1
∑i=1 (Xi − X̄)2 + ∑ j=1
2
(Y j − Ȳ )2 + ∑k=1
3
(Uk − Ū)2
(c) .
n1 + n2 + n3 − 2
n1
∑i=1 (Xi − X̄)2 + ∑nj=1
2
(Y j − Ȳ )2 + ∑nk=1
3
(Uk − Ū)2
(d) .
n1 + n2 + n3 − 3
S OLUTION
(a), (d).

E ND OF PAPER

11