ECO 329 (Economic Statistics) Midterm #2

Fall 2023 Semester

Professor Jason Abrevaya

This exam has a total of 40 points, with 4 additional points possible for extra credit.

There are 22 questions, worth 2 points each.

Exam time is 90 minutes. You should have sufficient time (4.09 minutes per question) to finish. Do not
spend a lot of time on any single question --- if you are stuck, move on to the next question. Please
leave the extra-credit questions for last.

Electronic devices are not permitted. The exam is closed-book, except for the provided “cheat sheet.”

Every question is either multiple choice or numerical answer.

• For multiple choice questions: Clearly circle one answer. No partial credit.

• For numerical answer questions: Clearly write your answer on the line provided, and give your
answer as a decimal or simplified fraction (unless otherwise specified). Use the space available
below the question to show your work, as partial credit may be awarded for some of these questions.

Student EID: _____________

Student name: __________________________________

1
PART I: Maria is a senior Economics major, and she is applying for jobs with consulting companies.
She applies to 30 different companies. If the probability of a job offer from any given company is 10%
and the companies can be treated as independent, the random variable associated with Maria’s total
number of job offers, denoted by !, is a Binomial(30, 0.1).

1. Which of the following probabilities is largest? (Hint: Think about the shape of the pmf graph
and where it peaks. You do not need to calculate Binomial probabilities for this question.)

0a. "(2 ≤ ! ≤ 4) pmf peaks3

b. "(6 ≤ ! ≤ 8) 30 o 11
at near
c. "(28 ≤ ! ≤ 30)

2. What is the variance of !? Give an exact numerical answer.

Answer: ________
2 7

1 0.9
Var Y nT I T 3010 2.7

3. Let ,! , ," , … , ,#$ denote the underlying i.i.d. Bernoulli(0.1) random variables associated with !.
That is, ! = ,! + ," + ⋯ + ,#$ . Now consider the random variables "! and "" , where.
!
"! = (,! + ," + ⋯ + ,"$ ) is the proportion of the first 20 companies that make an offer and
"$
!
"" = !$ (,"! + ,"" + ⋯ + ,#$ ) is the proportion of the last 10 companies that make an offer.
Which of the following statements is true?

a. The mean of "! is less than the mean of "" .

b. The standard deviation of "! is half as large as the standard deviation of "" .
!
Oc. The standard deviation of "! is
√"
times the standard deviation of "" .

!
d. The standard deviation of "" is times the standard deviation of "! .
√"

sd x proportional to
I

2
Maria decides to also apply for jobs with 30 finance companies, with the random variable 2 representing
the total number of finance job offers: 2 ∼ 4567859:(30, ;). Assume that job offers are independent
across all 60 companies (30 consulting, 30 finance).

4. Thinking about the proportion of job offers over the 60 companies (30 consulting, 30 finance),
!
represented by " = &$ (! + 2), what is the population average of "? (Your answer should be a
formula in terms of ;.)

t
or
Answer: ___________________
similar expression

Y Fl ELY
ELP E
LIL
3
1 301T

O.li

5. If ; = 0.1, what is the distribution of ! + 2?

a. 4567859:(30,0.2)
Bernoulli co
1 trials
d
Ob. 4567859:(60,0.1) Go i i
c. 4567859:(60,0.2)

d. None of the above

3
For two other senior Economics major, Jackson and Talia, the number of job offers are independent
draws of a random variable >, whose probability distribution is given by the following table:

T (job offers) 1 2 3 4
Probability 0.2 0.4 0.2 0.2

That is, Jackson will have between 1 and 4 job offers (inclusive), and Talia will have between 1 and 4
job offers (inclusive), each with the probabilities provided in the table.

6. Jackson’s starting salary will be 50 + 5>, in thousands of dollars. (The salary is an increasing
function of > since more offers provides more negotiating power.) What is the expected value of
Jackson’s starting salary? Give an exact numerical answer.

62
Answer: ___________

E T 1 0.2 2 0 Y 3 0.2 4 0.2 2.4

ST So SE IT 50 5712 4 62
EGO

!
Let @ be the average number of job offers received by Jackson and Talia. That is, @ = " (>! + >" ),
where >! is the number of offers for Jackson and >" is the number of offers for Talia.

7. How many possible outcomes are there for @?

7
Answer: ________

t 1.5 2 2.5 3 3 5,4

Outcomes

8. Thinking about the sampling distribution of @, what is the probability that @ = 3.5? Give an
exact numerical answer.

0.08
Answer: ________

4 P T 4,72 3
P A 3 5 P t 3 Tz

PCT 3 PCI 4 PCT 4 PCT 3

2
o 2 0.2 10.2710
o 08

4
PART II: (Normal running) Suppose that the marathon completion time (in minutes) of runners from a
certain Austin running club can be considered as draws of a normal random variable, > ∼ A(180,20" ).

For two runners on the club, assume that their running times are i.i.d. draws from this distribution, with
their times being represented by the random variables >! and >" .

9. What is B(>! )?

180
Answer: ________

10. The difference >! − >" is a normal random variable with …

mean = __________
0 800
and variance = ____________.

(fill in the blanks with exact numerical answers)

E T Ta E T E Tz I 80 180 0

Var Var Tz 202 202 800

Var Ti Ta ti

!
11. What is the variance of the average of the two times, given by " (>! + >" )? Give an exact
numerical answer.

Answer: ________________
200

Var ECT Ta I I 123 200

VEIT VEE
12. If the two runners’ times are not independent, but instead have a positive covariance (D'! '" > 0),
how would the variance of the average of their times compare to your answer from the previous
question?

O
a. The variance of the average is higher when there is positive covariance between >! and >" .

b. The variance of the average is lower when there is positive covariance between >! and >" .

c. The variance of the average is the same when there is positive covariance between >! and >" .

5
Now suppose that there are 100 runners from the Austin running club, with times given by i.i.d. draws
>! , >" , ># , … , >!$$ from the > ∼ A(180,20" ) distribution. Let >F denote the average of the 100 times.

13. What is B(>F)?

T is normal
O
a. 180
with ECE 180
!($ Mi
4
3
b. !$$
and F rare 0
c. Can’t tell without additional information

14. Which of the following is statements is true?

T and T both normal
a. "(179 ≤ >F ≤ 181) = "(179 ≤ >! ≤ 181) mean 180
with same
a
b. "(179 ≤ >F ≤ 181) > "(179 ≤ >! ≤ 181)
O I having
c. "(179 ≤ >F ≤ 181) < "(179 ≤ >! ≤ 181) lower variance

15. What is the 95% quantile of >F? Your answer should only be in terms of numbers, but you do not
need to simplify your answer. (Hint: Use the fact that >F is normally distributed.)

Answer: ________________________
T
quantile of
957 is
T N 1801,4
1.645 z
180
2
M F F
EC1. (Do not attempt until you have completed the rest of the exam) The first runner from the
Austin club, whose time is >! ∼ A(180,20" ), is racing against another runner (Bob) whose time
is described as a discrete random variable 4 with "(4 = 140) = 0.4 , "(4 = 180) = 0.4, and
"(4 = 220) = 0.2. Assuming the two runners’ times are independent, what is the probability
that the first runner beats Bob (>! < 4)? (For this question, you may assume that the value “2”
(rather than 1.96) gives you exactly 2.5% in each tail of a normal distribution.) You must give
the exact numerical answer, as a decimal, to receive the extra credit.

0.405
Answer: ____________

B P B 140 PCT B B 140

PCT
PCB 180 P T B B 180
B B 220
P B 2207 PIT
0.025 0.4 o 5 O 2 0.9753
o 4

o o f t o 2 t o 195 0.405 6
PART III: (Non-normal running) Suppose that the marathon completion time (in minutes) of runners
from a certain Houston running club can be considered as draws of a random variable , whose pdf is:

0.012 5J 130 ≤ K ≤ 180

J) (K) = L0.010 5J 180 < K ≤ 220
0 7NℎPQR5SP

16. What is "(, = 180)?

a. 0.012

b. 0.010

Oc. 0
17. What is the cdf of , evaluated at 120? (That is, what is 2) (120)?) Give an exact numerical
answer.

O
Answer: ________

F 7207 P XE 1207 0

(Hint: For the next two questions, it may be helpful to sketch a graph of the pdf J) (K) if you are stuck.)

18. The median of , is:

180 P XE 180
Fx
a. Less than 180 012
180 130 0
b. Equal to 180
0.6
c. Greater than 180
is below 180
median

19. What is "(160 ≤ , ≤ 200)? Give an exact numerical answer.

0.44
Answer: ________

P IGOE X 180 P 180C XE 200

P 1601 4200
20 0.010
20710.012
0.24 t o 20 0.44

7
20. It can be shown that the mean of , is equal to 173 and the standard deviation of , is 26. Letting
)
! = &$ denote the marathon completion time in hours, which of the following statements is true?

"&

If
a. The mean and standard deviation of ! are 173 and , respectively.
My
&$

!*# "&

0b. The mean and standard deviation of ! are &$

and &$, respectively.

c. The mean and standard deviation of ! are

!*#
&$
"&
and #&$$, respectively. r E
EC2. (Do not attempt until you have completed the rest of the exam) Suppose that a Houston club
runner (described by the random variable , at the beginning of Part III) is racing against another
runner (Stacy) whose completion time is uniformly distribution, with T ∼ U(170, 190).
Assuming the two runners’ times are independent, what is the probability that the Houston
runner beats Stacy (, < T)? You must give the exact numerical answer, as a decimal, to receive
the extra credit. (The question will require multiplying decimals together, so simplify until you
get a single number as your answer.)

Here is the pdf of , again so that you don’t need to flip back to the previous page:

0.012 5J 130 ≤ K ≤ 180

J) (K) = L0.010 5J 180 < K ≤ 220
0 7NℎPQR5SP

0.595
Answer: ___________
05
f x 10 0

190 TIE 170,1907

i Yao

of P
P X CS joint prob of D t j int prob
05 is 10 0.012 10.05
207140 0.012710
5 IO 0.010 o.o

0006
800710 150710.0006 50 o 0005

0.48 0.09 0.025

8
0.595