Econometrics exam

Midterm 1

ECO 231 - Undergraduate Econometrics

Prof. Kegon Tan

INSTRUCTIONS

The instructions below are the same that will be given to you in the exam.

1. This exam has 2 questions. First question will add up to 70 points, while the second

question is worth 30 points. Each item inside a question is worth the same.

2. You are not allowed the use of notes, cheat sheets, calculators, or electronic devices

of any kind. Turn your cell phone off, and put it away. The TAs will announce the

time every 15 minutes. If your answers are unclear or illegible you may lose points.

3. Keep your video camera on at all times during the examination.

4. Submit all written work through Blackboard.

5. Please join the zoom room 5 minutes before the start of the exam.

6. Exam format: Your exam must be delivered as a Word or PDF file. You can type

it or you can write your answers directly in your laptop/tablet. Some parts of the

exam require you to write equations/graphs. You are allowed to submit ONE picture

of ONE SINGLE PAGE for your answers to those questions. This means that if you

type your exam in Word but decide to make the graphs/equations handwritten, you

can submit ONE PICTURE of ONE SINGLE PAGE with those answers along with

the Word file. If you decide to write it with a pen in your iPad/laptop/tablet you can

either do it directly in your main Word of PDF document, or submit it separately as

a one SINGLE PAGE document.

7. Once the exam time is over, you will have 5 minutes to submit all your files (main

Word plus additional file with handwritten material if necessary) to Blackboard. We

will close the submission portal once the time is over.

1

1 Material Question - 7 points each

A city has implemented an after school program to provide students homework help and

extra learning. The program is free, not mandatory and students are free to attend or not.

However, schools do encourage student with low academic performance to take advantage

of the program by attending it.

The city mayor wants to assess whether the program has helped to improve student’s

grades. To be specific, program attendance is measured as the total number of hours the

student attended the program in the school year. Grade is measured as the student’s

average grade in test scores taken during the school year. Call the variable “after school

program” as pr and the variable “grade” as grd.

1. Describe the ideal experiment to answer this question.

2. Suppose that I want to estimate the after school program’s (pr) effect on grades

(grd). Consider the univariate regression line:

grd = a + b1pr

how to calculate a and b1 using OLS?

3. Given the context in which program is implemented, would you expect the coefficient

b1 to be positive or negative? Explain why.

4. Is the variable that measures how much time the student spends playing video games

a confounder? The variable vd is measured as the total number of hours spent playing

video game during the school year. Explain your answer.

5. Suppose that the estimated regression line is grd = 85 − 0.12 ·pr. Comment on the following sentence: if pr increases by two additional hours, then the predicted grade

will decrease by 0.24.

6. Suppose that my data set contains students’ grades (grd), hours in the after school

program (pr), hours spent playing video game (vd), and student’s family income

(inc). Consider the multivariate regression line:

grd = a + b1pr + b2vd + b3inc

Express the partialling-out formula of the regression coefficient b1 in terms of the

variables in the model. Interpret it.

2

7. Suppose that the model is

grd = β0 + β1pr + β2vd + β3inc + u

where E[u|pr,vd,inc] = 0. What is this model saying about the world?

8. Interpret β0 and β1 in this model.

9. Do you believe that, after controlling for family income, b1 (pr’s coefficient) is smaller,

greater or equal to b1 from item (2)? Explain your answer.

10. How do you expect the R2 to change if we additionally control for parental education?

Why?

3

2 Paper Question - 6 points each

This question refers to the paper titled “Home advantage in European international soccer:

which dimension of distance matters?” (Nils Van Damme and Stijn Baert, 2019).

• Relevant tables and figures are included below.

• Approximate calculations are fine since you do not have a calculator.

1. What is the aim of the paper, and who would be interested in the conclusions of the

authors?

2. Explain why multiple distance measures are needed in the analysis of the paper.

Discuss at least two of them in your explanation.

3. Interpret the coefficient β for Di,n in Table 2. How much is the overall home advantage

for goal difference, number of points and victory respectively?

4. Interpret and compare the R2’s in Table 2.

5. Explain why the coefficient estimates in this paper cannot be interpreted as a causal

relationship. Suggest a way to overcome this limitation in an ideal research setting.

4

2 .J

P G

T a b

le 2

5