ECON7310: Elements of Econometrics
Research Project 2
Fu Ouyang
September 20, 2021
Instruction
Answer all questions following a similar format of the answers to your tutorial questions. When
you use R to conduct empirical analysis, you should show your R script(s) and outputs (e.g.,
screenshots for commands, tables, and figures, etc.). You will lose 2 points whenever you fail to
provide R commands and outputs. When you are asked to explain or discuss something, your
response should be brief and compact. To facilitate tutors’ grading work, please clearly label
all your answers. You should upload your research report (in PDF or Word format) via the
“Turnitin” submission link (in the “Research Project 2” folder under “Assessment”) by 12:00
PM on the due date October 5, 2021. Do not hand in a hard copy. You are allowed to work
on this assignment in groups; that is, you can discuss how to answer these questions with your
group members. However, this is not a group assignment, which means that you must answer
all the questions in your own words and submit your report separately. The marking system will
check the similarity, and UQ’s student integrity and misconduct policies on plagiarism apply.
A. Panel Data (60 points)
Background
Do citizens demand more democracy/political freedom as their incomes grow? That is, is democ-
racy a normal good? To study this issue, you will explore the dataset income democracy.csv
which contains a panel data set from 195 countries for the years 1960, 1965, …, 2000. A
detailed data description is given in the table below.1 The dataset contains an index of democ-
racy/political freedom for each country in each year, together with data on the country’s income
and various demographic controls. The income and all demographic controls are lagged five
years relative to the democracy index to allow time for democracy to adjust to changes in these
variables.
Research Questions
1. (3 points) Is the data set a balanced panel? Explain. Hint: Look through the dataset
or use the is.pbalanced() function.
2. (11 points) The index of political democracy/political freedom is labeled dem ind.
(a) What is the value of dem ind for the United States in 2000 (1 point)? What is the
average of dem ind for the United States over all years in the data set (2 points)?
Repeat this exercise for Libya (2 points).
1These data were provided by Daron Acemoglu of M.I.T. and were used in his paper with Simon Johnson,
James Robinson, and Pierre Yared, “Income and Democracy,” American Economic Review, 2008, 98:3, 808–842.
1
(b) How many missing values are there for dem ind (1 point)? List all countries with the
lowest average value of dem ind (i.e., =0) (4 points). How many countries do have
the highest average value of dem ind (i.e., =1) (1 point)? Hint: You can identify
missing values using the is.na() function and compute sample average for each
country using the aggregate() function.
3. (16 points) The logarithm of per capita income is labeled log gdppc.
(a) Regress dem ind on log gdppc using standard errors that are clustered by country.
Report your estimation results in a table similar to Table 10.1 on page 378 of SW
textbook, not including the test part (3 points).
(b) Interpret the estimated coefficient on log gdppc (2 points)? Is the coefficient statis-
tically significant (1 point)?
(c) If per capita GDP in a country increases by 10%, by how much is dem ind predicted
to increase (2 points)? Construct a 95% confidence interval for the prediction (2
points)? Is the predicted increase in dem ind large or small? Explain what you mean
by large or small (2 points).
(d) Why is it important to use clustered standard errors for the regression (2 points)?
Do the results change if you do not use clustered standard errors (2 points)?
4. (27 points) Consider panel data regressions.
(a) Suggest a variable that varies across countries but plausibly varies little (or not at
all) over time and that could cause omitted variable bias in the regression in Question
3 (Q3) above (3 points).
(b) Estimate the regression in Q3, allowing for country fixed effects. Add the estimation
results to the table in 3(a) (3 points). How do your answers to 3(b) and 3(c) change
(2 points)?
(c) Exclude the data for Azerbaijan and re-run the regression. Do the results change (2
points)? Why or why not (2 points)?
(d) Suggest a variable that varies over time but plausibly varies little (or not at all)
across countries and that could cause omitted variable bias in the regression in Q3
(3 points).
(e) Estimate the regression in Q3, allowing for both time and country fixed effects. Add
the estimation results to the table in 3(a) (3 points). How do your answers to 3(b)
and 3(c) change (4 points)?
2
(f) There are additional demographic controls in the data set. Should these variables
be included in the regression? If so, re-run the regression, including these controls.
Report the regression results in the table in 3(a) (3 points). How do the results
change when they are included (2 points)?
5. (3 points) Based on your analysis, what conclusions do you draw about the effects of
income on democracy?
B. Binary Choice Model (40 points)
Background
The Australian government passed its $130 billion JobKeeper package to reduce the unemploy-
ment rate during the lockdown. But, it seems that the longer this lasts, the higher the numbers
of unemployment goes. You want to study what kind of workers might lose their jobs during
the COVID-19 health crisis. Specifically, are young workers more likely to lose their jobs than
middle-aged workers? Does their educational background help them to keep their jobs?
Unfortunately, you have only one dataset (i.e., US Employment 0809.csv), which contains a
random sample of 5,412 U.S. workers. They were surveyed in 2008 after the global financial crisis
of 2007–2008 and reported that they were employed. The data file contains their employment
status (i.e., employed or not) in 2009, one year later, along with some additional variables. A
detailed data description is given in the following table:
Research Questions
(a) (10 points) Using a linear probability model (LPM) to regress employed on age and
age2 and report regression results (5 points). Is age a statistically significant determinant
3
of employment (3 points)? Is there evidence of a nonlinear effect of age on the probability
of being employed (2 points)?
(b) (10 points) Repeat (a) using a probit and logit regression model and report your results
(6 points). Explain the difference between probit and logit model (4 points).2
(c) (8 points) Use the models of LPM, probit, and logit to compute the predicted probability
of employment for a 20-year-old worker, a 40-year-old worker, and a 60-year-old worker
and fill the following table (5 points). Explain the differences in the results (3 points).
Table 1: Predicted Probability
LPM Probit Logit
̂P (employed = 1|age = 20) 0.730
̂P (employed = 1|age = 40) 0.916 0.911
̂P (employed = 1|age = 60) 0.832
(d) (12 points) The data contains variables measuring the worker’s educational attainment,
gender, race, marital status, region of the country, and weekly earnings. Repeat (a) and
(b) using these factors as additional regressors and construct a table (such as Table 11.2 in
SW textbook, not including the test part on page 411) (6 points). Investigate whether the
conclusions on the effect of age on employment from (a) and (b) are reliable or not when
including these control variables (3 points). Use the regressions in your table to discuss
the characteristics of workers who were hurt most by the financial crisis (3 points). Hint:
Include dummies for race groups and ln(earnwke+ 1).3
2As probit and logit assume homoskedasticity, you don’t need to compute robust SE.
3We do this transformation to avoid the presence of ln(0) = −∞.
4