ECONOMICS 2123B

MIDTERM 1 (SOLUTION)
2021-01-27 TO 2021-01-29

INSTRUCTOR: RENFANG TIAN

Name:
Student ID:

Instructions: There are 40 multiple choice questions, and 5 short answer questions. For each multiple
choice question, please select one of A, B, C or D. If you select more than one answer, no marks will be
awarded for that question. For short answer questions, please show all your work. Each multiple choice
question is worth 2 marks and each short answer question is worth up to 4 marks each.

Multiple Choice
(1) Econometrics can be described as follows with the exception of
A) the science of testing economic theory.
B) fitting mathematical economic models to real-world data.
C) a set of tools used for analyzing relationships among economic variables.
D) measuring room temperatures.

(2) One of the primary advantages of using econometrics is

A) it potentially provides you with quantitative answers based on data for practical prob-
lems rather than simply suggesting the direction (positive/negative) of the response.
B) teaching you how to use statistical packages.
C) learning how to invert a 4 by 4 matrix.
D) all of the above.

(3) To provide quantitative answers to policy questions

A) it is typically sufficient to use common sense.
B) you should interview the policy makers involved.
C) you should examine empirical evidence using quantitative analysis.
D) is typically impossible since policy questions are not quantifiable.

1
ECON 2123B Midterm 1 (Solution) Instructor: Renfang Tian

(4) An econometric model is

A) an equation.
B) a set of random variables.
C) a set of restrictions that specify relationships among variables.
D) an estimation.

(5) We estimate econometric models

A) based on our believes about the relationship among variables.
B) using estimators and observed data.
C) because we want to learn about the phenomena or the relationships in the population level.
D) all of the above.

(6) An estimator is
A) a variable.
B) a rule for obtaining the estimate.
C) a function of the random sample.
D) all of the above.

(7) A sample
A) always shares the same distribution as the population.
B) is a discrete random variable.
C) is a subset of a population.
D) is all of the above.

(8) Suppose we are interested in the relationship between hours of study and students’ performance in
tests for Ontario’s high school students, then our population is
A) All the Ontarians.
B) All the Ontario’s high school students.
C) All the Ontario students.
D) All the students in the high school close to your residence.

(9) We are interested in the relationship between hours of study and students’ performance in tests for
Ontario’s high school students. Which procedure is the most likely to provide a random sample?
A) Collecting data from one or two high schools close to your residence.
B) Listing all the Ontario high schools, randomly picking 10 of them, and Collecting data from all
grade-12 (last-year) students in those 10 selected high schools.
C) Collecting data from all the high schools in London.
D) Listing all the Ontario high schools, randomly picking 10 of them, and Collecting
data from random students in those 10 selected high schools.

2
 ECON 2123B Midterm 1 (Solution) Instructor: Renfang Tian

(10) Random sampling has the following properties except that

A) each member is selected by chance.
B) it gives equal chance to all the population members to be selected in a sample.
C) generating sample sets in which every value has an equal probability to occur.
D) the selection of each member does not affect the selection of others.

(11) A random variable is

A) a fixed number.
B) a function.
C) a random outcome.
D) all of the above.

(12) The behaviour of a random variable can be described by

A) its CDF.
B) its PDF or PMF.
C) its expectation or variance.
D) all of the above.

(13) Which of the following is NOT a random variable?

A) The number of vehicles passing a traffic light in the next hour.
B) The side (Head or tail) you get from tossing a coin.
C) The revenue an automobile store makes tomorrow.
D) The number of cars an automobile store sales tomorrow.

(14) Which of the following is a continuous random variable?

A) The number of vehicles passing a traffic light in the next hour.
B) The side (Head or tail) you get from tossing a coin.
C) The revenue an automobile store makes tomorrow.
D) The number of cars an automobile store sales tomorrow.

(15) A discrete random variable

A) is one that obtained from discrete observation.
B) is one that takes on only countably many values.
C) must have a continuous CDF.
D) must have a constant PMF.

3
 ECON 2123B Midterm 1 (Solution) Instructor: Renfang Tian

(16) A PMF
A) can provide the number of times a basic outcome occurs in random experiments.
B) equals to the number of times a basic outcome occurs divided by the number of times the random
experiment repeats.
C) is used to indicate the probability function of a discrete random variable or a continuous random
variable.
D) gives the probability that a discrete random variable is equal to some value.

(17) A PDF
A) equals to the number of times a basic outcome occurs divided by the number of times the random
experiment repeats.
B) is used to indicate the probability function of a discrete random variable or a continuous random
variable.
C) gives the probability that a continuous random variable is equal to some value.
D) gives the density that a continuous random variable is equal to some value.

(18) A CDF
A) is only for discrete random variables.
B) is only for continuous random variables.
C) gives the probability that a random variable is less than or equal to a particular
value.
D) can provide the number of times a basic outcome occurs in random experiments.

(19) A parameter can

A) describe a characteristic of a population.
B) describe a characteristic of a sample.
C) always be observed.
D) all of the above.

(20) A statistic can

A) be computed from a sample.
B) describe a characteristic of a sample.
C) be a random variable.
D) all of the above.

4
 ECON 2123B Midterm 1 (Solution) Instructor: Renfang Tian

(21) The expected value of a discrete random variable

A) is the outcome that is most likely to occur.
B) can be found by determining the 50% value in the c.d.f.
C) is always an integer.
D) is computed as a weighted average of the possible outcome of that random variable,
where the weights are the probabilities of that outcome.

(22) An expectation has the property/properties

A) E[XY ] = E[X]E[Y ] always, for random variables X and Y .
B) E[a] = 0 for a constant a.
C) E[X + Y ] = E[X] + E[Y ] always, for random variables X and Y .
D) E[aX] = aX for a constant a and a random variable X.

(23) A sample average

A) is a random variable and has a probability distribution.
B) is a single number and as a result cannot have a distribution.
C) equals to expectation.
D) needs to be computed using PMF.

(24) A population variance

A) equals to the squared expectation.
B) is the square root of the population standard deviation.
C) measures the dispersion of the distribution of a random variable.
D) measures how far the random variable deviates from zero.

(25) A variance has the property/properties

A) Var(X) = E[X 2 ] − (E[X])2 for a random variable X.
B) Var(aX) = a Var(X) for a constant a and a random variable X.
C) Var(X − Y ) = Var(X) + Var(Y ) always, for random variables X and Y .
D) Var(X − Y ) = Var(X) − Var(Y ), if X and Y are uncorrelated random variables.

(26) A sample variance

A) can be computed with or without knowing the expectation.
B) describes the dispersion of the sample.
C) is a statistic.
D) all of the above.

5
 ECON 2123B Midterm 1 (Solution) Instructor: Renfang Tian

(27) When interpreting a covariance,

A) it’s important to look at the +/– sign.
B) it’s important to look at the magnitude.
C) it indicates the strength of the relationship.
D) all of the above.

(28) A covariance has the property/properties

A) Cov(X, X) = Var(X) for a random variable X.
B) Cov(X − X, Y ) = 0 for random variables X and Y .
C) Cov(2X, Y ) = Cov(X, 2Y ) for random variables X and Y .
D) all of the above.

(29) When interpreting a correlation coefficient,

A) it’s important to look at the +/– sign.
B) it’s important to look at the magnitude.
C) it indicates the strength of the relationship.
D) all of the above.

(30) The correlation between two random variables X and Y

A) cannot be negative since variances are always positive.
B) is the covariance squared.
C) can be calculated by dividing the covariance between X and Y by the product of the
two standard deviations.
Cov(X,Y )
D) is given by corr(X, Y ) = Var(X) Var(Y ) .

(31) If two random variables are positively correlated,

A) their covariance will be very large.
B) their covariance will be positive.
C) we cannot tell from the covariance.
D) their correlation coefficient will equal to 1.

6
 ECON 2123B Midterm 1 (Solution) Instructor: Renfang Tian

(32) Assume that a student assigns the following subjective probabilities to the possible outcome in a
course (see Table 1): The expected GPA value is

Grade GPA Probability

A 4 0.20
B 3 0.50
C 2 0.20
D 1 0.08
F 0 0.02
Table 1. GPA

A) 2.00.
B) 2.78.
C) 3.00.
D) 3.50.

(33) According to the probabilities in Table 1, the variance of the GPA value is
A) 0.85.
B) 0.92.
C) 2.50.
D) 2.00.

(34) According to the probabilities in Table 1, the probability of getting A or B is

A) 0.70.
B) 0.20.
C) 0.50.
D) uncertain.

(35) Two random variables X and Y are independently distributed if

A) knowing the value of either one does not affect the distribution of the other.
B) knowing the value of either one provides very little information about the other.
C) E[XY ] = 0.
D) E[XY ] = E[X]E[Y ].

(36) Two random variables X and Y are uncorrelated if

A) knowing the value of either one does not affect the distribution of the other.
B) E[XY ] = 0.
C) E[XY ] 6= E[X]E[Y ].
D) E[Y ] = 0

7
 ECON 2123B Midterm 1 (Solution) Instructor: Renfang Tian

(37) Two random variables are IID if

A) they are uncorrelated.
B) they have the same variance.
C) if they are independent and have the same distribution.
D) if the realization of one does not affect the distribution of the other.

(38) An estimator β̂ for β is unbiased if

A) β̂ = β.
B) Var(β) = 0.
C) E[β] = β̂.
D) E[β̂] = β.

(39) Given a random sample X1 , ..., Xn that are IID, which of the following estimators µ̂ for the expecta-
tion E[X] = µ is unbiased?
A) µ̂ = X1 .
B) µ̂ = n1 ni=1 Xi .
P

C) µ̂ = 21 (X1 + Xn ).
D) all of the above.

2 2
(40) For the two estimators σ̂ = n1 ni=1 (Xi − X̄) and σ̃ = n−1
1 Pn
P  
i=1 (Xi − X̄) for the population
variance σ ,2

A) σ̂ outperforms σ̃ in terms of estimation accuracy, on average.

B) σ̃ outperforms σ̂ in terms of estimation accuracy, on average.
C) σ̃ is always closer to σ 2 than σ̂.
D) σ̂ is always closer to σ 2 than σ̃.

8
ECON 2123B Midterm 1 (Solution) Instructor: Renfang Tian

Short Answers
(1) To infer the political tendencies of college/university students in London Ontario, propose a feasible
sampling procedure that can provide a random sample of size n = 100.
Solution: There are many different aspects to consider to ensure the randomness of our sample. For
example, if we propose to divide the population into subgroups first and then collect samples from
each subgroup, then we need to take into consideration of the proportion each subgroup takes out of
the entire population. In practice, we also need to consider whether certain subgroups would tend to
not respond.

(2) Consider two stocks A and B with the prices shown as follows. Denote the stock prices as X and Y

Stock A Stock B
2.5 2.0
2.2 2.8
0.8 1.4
1.5 2.0
2.4 3.0
Table 2. Stock Prices

for stocks A and B respectively. Using the data for X and Y in Table 2, estimate the expectations
for X and Y .
Solution: We use the sample averages (2 points) to estimate the expectations.
n
1X 1
X̄ = Xi = (2.5 + 2.2 + 0.8 + 1.5 + 2.4) = 1.88 (1 point);
n 5
i=1
n
1X 1
Ȳ = Yi = (2.0 + 2.8 + 1.4 + 2.0 + 3.0) = 2.24 (1 point)
n 5
i=1

(3) Using the data for X and Y in Table 2, estimate the variances for X and Y and interpret your result.
Solution: Since the sample size is small and we do not know the expectation, we can use the adjusted
9
ECON 2123B Midterm 1 (Solution) Instructor: Renfang Tian

sample variances (2 points) to estimate the population variances.

n
2 1 X
σ̂X = (Xi − X̄)2
n−1
i=1
1
= [(2.5 − 1.88)2 + (2.2 − 1.88)2 + (0.8 − 1.88)2 + (1.5 − 1.88)2 + (2.4 − 1.88)2 ]
4
= 0.517 (0.5 point);
n
1 X
σ̂Y2 = (Yi − Ȳ )2
n−1
i=1
1
= [(2.0 − 2.24)2 + (2.8 − 2.24)2 + (1.4 − 2.24)2 + (2.0 − 2.24)2 + (3.0 − 2.24)2 ]
4
= 0.428 (0.5 point)

The sample variance estimates the dispersion of the two stock prices, which indicates that Stock A
is more risky than Stock B (1 point).

If you treat the data as the population data, then use

n
2 1X
σ̂X = (Xi − E[X])2
n
i=1
1
= [(2.5 − 1.88)2 + (2.2 − 1.88)2 + (0.8 − 1.88)2 + (1.5 − 1.88)2 + (2.4 − 1.88)2 ]
4
= 0.4136 (0.5 point);
n
1X
σ̂Y2 = (Yi − E[Y ])2
n
i=1
1
= [(2.0 − 2.24)2 + (2.8 − 2.24)2 + (1.4 − 2.24)2 + (2.0 − 2.24)2 + (3.0 − 2.24)2 ]
4
= 0.3424 (0.5 point)

(4) Using the data for X and Y in Table 2, estimate the covariance between X and Y and interpret your
result.
Solution: Since the sample size is small and we do not know the expectation, we can use the adjusted
sample covariance (2 points) to estimate the population covariance.
n
1 X
σ̂XY = (Xi − X̄)(Xi − X̄)
n−1
i=1
1
= [(2.5 − 1.88)(2.0 − 2.24) + (2.2 − 1.88)(2.8 − 2.24) + (0.8 − 1.88)(1.4 − 2.24)+
4
(1.5 − 1.88)(2.0 − 2.24) + (2.4 − 1.88)(3.0 − 2.24)]
= 0.356 (1 point).

10
ECON 2123B Midterm 1 (Solution) Instructor: Renfang Tian

since σ̂XY = 0.356 > 0, the estimated covariance indicates that the two stock prices are positively
correlated (1 point).

If you treat the data as the population data, then use

n
1X
σ̂XY = (Xi − E[X])(Xi − E[X])
n
i=1
1
= [(2.5 − 1.88)(2.0 − 2.24) + (2.2 − 1.88)(2.8 − 2.24) + (0.8 − 1.88)(1.4 − 2.24)+
5
(1.5 − 1.88)(2.0 − 2.24) + (2.4 − 1.88)(3.0 − 2.24)]
= 0.2848 (1 point).

(5) Using the data for X and Y in Table 2, estimate the correlation coefficient between X and Y and
interpret your result.
Solution: Since the sample size is small and we do not know the expectation, we can use the adjusted
sample covariance (2 points) to estimate the population covariance.
σ̂XY
ρ̂XY = q
2 σ̂ 2
σ̂X Y
0.356
=√ ≈ 0.7568 (1 point).
0.517 ∗ 0.428
since ρ̂XY ≈ 0.7568 > 0, which is relatively close to 1, the estimated correlation coefficient indicates
that the two stock prices are strongly positively correlated (1 point).

If you treat the data as the population data, then use

σ̂XY
ρ̂XY = q
2 σ̂ 2
σ̂X Y
0.2848
=√ ≈ 0.7568 (1 point).
0.4136 ∗ 0.3424

11