EC2C4: ECONOMETRICS II

PROBLEM SETS

• Questions marked with the * symbol will typically be covered with priority in class.

• Try to solve as many questions as you can before your class. Nevertheless, do not worry
if you cannot solve everything by yourself. You are welcome to request your teachers to
help you (during class and/or office hours) with the parts that you find difficult.

• The so-called “Lecture Review” question at the end will NOT be discussed in class.
There will be no separate answer provided for it either. It is just a prompt to encourage
you to go back and review the lecture notes.

• Please prepare Problem Set 1 below for class in Week 2.

1
 Problem set 1

(1) [Wooldridge Ch. 2, Problem 4] The dataset [Link] contains data on births
to women in the United States. Two variables of interest are the dependent variable,
infant birth weight in ounces (bwght), and an explanatory variable, average number of
cigarettes the mother smoked per day during pregnancy (cigs). The following simple
regression was estimated using data on n = 1, 388 births:
\ = 119.77 − 0.514cigs
bwght

(i) What is the predicted birth weight when cigs = 0? What about when cigs = 20
(one pack per day)? Comment on the difference.
(ii) Does this simple regression necessarily capture a causal relationship between the
child’s birth weight and the mother’s smoking habits? Explain.
(iii) To predict a birth weight of 125 ounces, what would cigs have to be? Comment.
(iv) The proportion of women in the sample who do not smoke while pregnant is about
0.85. Does this help reconcile your finding from part (iii)?

(2) * [Wooldridge Ch. 2, Problem 6] Using data from 1988 for houses sold in Andover,
Massachusetts, from Kiel and McClain (1995), the following equation relates housing
price (price) to the distance from a recently built garbage incinerator (dist):
\ = 9.40 + 0.312 log(dist)
log(price)

n = 135, R2 = 0.162
(i) Interpret the coefficient on log(dist). Is the sign of this estimate what you expect
it to be?
(ii) Do you think simple regression provides a good estimator of the ceteris paribus
elasticity of price with respect to dist? (Think about the city’s decision on where
to put the incinerator.)
(iii) What other factors about a house affect its price? Might these be correlated with
distance from the incinerator?

2
(3) * [Wooldridge Ch. 2, Problem 12] Consider the problem of running a regression and
only estimating an intercept.
(i) Given a sample {yi : i = 1, 2, . . . , n}, let β̃0 be the solution to
n
X
min (yi − b0 )2 .
b0
i=1

Show that β̃0 = ȳ, that is, the sample average minimizes the sum of squared
residuals.
(ii) Define residuals ũi = yi − ȳ. Argue that these residuals always sum to zero.

(4) * Consider the bivariate regression yi = β0 + β1 xi + ui , which satisfies the condition that
the sample variance of x is strictly positive. Let β̂1 be the OLS slope estimate and R2
be the R-squared.
(i) Suppose β̂1 = 0. Show that R2 = 0.
(ii) Suppose R2 = 0. Does this imply β̂1 = 0?

(5) Let β̂0 and β̂1 be the intercept and slope from the regression of y on x. Let c1 and c2 be
constants with c2 6= 0. The following questions consider the effects of transformations,
y−c1 x−c1
such as c2
and c2
(a typical example is standardization of data).
y−c1
(i) Let β̃0 and β̃1 be the intercept and slope from the regression of c2
on x. Express
β̃0 and β̃1 in terms of β̂0 , β̂1 , c1 , and c2 .
x−c1
(ii) Let β̃0 and β̃1 be the intercept and slope from the regression of y on c2
. Express
β̃0 and β̃1 in terms of β̂0 , β̂1 , c1 , and c2 .
yi −c1
(iii) Now suppose yi > 0 and c2
> 0 for all i, and redefine β̂0 and β̂1 as the intercept
and slope from the regression oflog(y) on x. Let β̃0 and β̃1 be the intercept and
y−c1
slope from the regression of log c2 on x. Can we express β̃0 and β̃1 in terms
of β̂0 , β̂1 , c1 , and c2 ? Explain.

(6) [Lecture Review.] Please ensure that you know how to derive the method of moments
estimators of β0 and β1 in the simple linear regression model given on slide 22 of the
“Bivariate Regression” slide pack. (The answer is directly in the lecture slides.)
(i) In particular, pay special attention to each line of the algebra presented in slides
23–30. (For example, do you understand why Σni=1 (xi −x̄)(yi −ȳ) = Σni=1 xi (yi −ȳ))?
(ii) Ensure that you can explain the difference (if any) between method of moments
and ordinary least squares estimators both in terms of general methodology and
in this specific simple linear regression setting.