lOMoARcPSD|2960605

Exam June, questions

Basic Econometrics (Royal Melbourne Institute of Technology)

StuDocu is not sponsored or endorsed by any college or university

Downloaded by Quy?nh ???ng ([email protected])
 lOMoARcPSD|2960605

SECTION A
Answer the following multiple choice questions using the Multiple Choice Answer Sheet.
Each question is worth two marks. This section is worth a total of 40 marks.

QUESTION 1

In the equation y = β 0 + β 1x + u, the term y is the _____.

(a) dependent variable.

(b) independent variable.

(c) slope parameter.

(d) intercept parameter.

QUESTION 2

Consider the following simple regression model of house prices:

house_price = b0 + b1*land_size + u. What is b1?

(a) land_size.

(b) the distance to the city.

(c) slope parameter.

(d) intercept parameter.

QUESTION 3

Consider the following simple regression model of house prices:

house_price = β0 + β1*land_size + u. If the expected value of house_price is zero when
land_size is zero, β0 will be

(a) positive

(b) negative

(c) zero

(d) exactly twice as large as β1

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 1 of 10

Page 1 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

QUESTION 4

In the equation, y=β 0 + β 1 x 1 +β 2 x 2 +u, β 0 is a(n) _____.

(a) independent variable

(b) dependent variable

(c) slope parameter

(d) intercept parameter

QUESTION 5

Consider the regression model wage = β 0 + β 1 Age+u , where u is correlated with Age . Which
of the following assumptions is violated for sure:

(a) random sampling

(b) linear in parameters

(c) conditional mean independence

(d) no perfect multicollinearity

QUESTION 6

The value of R2 always _____.

(a) lies above 0.

(b) lies above 1.

(c) lies between -1 and 0.

(d) lies between 1 and 1.5.

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 2 of 10

Page 2 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

QUESTION 7

Suppose the following equation is estimated where sleep is measured in minutes and work
in hours:

sleep = 3638.25 – 14.8*work - 11.13*educ + 2.20*age

n=706, R2 =0.0111.

If someone gets one more year of education, by how many minutes is sleep predicted to
fall?

(a) 14.8.

(b) 11.13.

(c) 29.6.

(d) 11.13%.

QUESTION 8

Suppose the following equation is estimated:

sleep = 3638.25 – 14.8*work - 11.13*educ + 2.20*age

n=706, R2 =0.0111.

What percentage of the variation in sleep is explained by the regression?

(a) 1.11%.

(b) 11.1%

(c) 11.13%.

(d) 0.1113%.

QUESTION 9

The general confidence interval can be written as:

(a) estimate ±criticalvalue∗standard error

(b) estimate ±criticalvalue /standard error

(c) estimate /standard error

(estimate−hypothesised value)
(d)
standard error

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 3 of 10

Page 3 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

QUESTION 10

Consider the regression equation log(wage) = 0.284 + 0.092*educ + 0.0045 * exper. The
standard error of the educ coefficient estimate is 0.034. Assume the null hypothesis is that
the educ coefficient is equal to zero. What is the t-statistic for the educ coefficient?

(a) 0.092.

(b) 0.271.

(c) 0.370.

(d) 2.71.

QUESTION 11

Consider the equation log(unemployed) = 0.284 - 0.025*spending + 0.0045 * int_rate. By

how much (approximately) does unemployed decrease when spending increases by 2?

(a) 0.50%.

(b) 2.5%

(c) 5%.

(d) 0.25%.

QUESTION 12

Consider the equation log(unemployed) = 0.284 - 0.025*spending + 0.0045 * int_rate. By

how much (EXACTLY) does unemployed decrease when spending increases by 2?

(a) 5.1%.

(b) 5.2%

(c) 0.052%.

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 4 of 10

Page 4 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

(d) 0.051%.

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 5 of 10

Page 5 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

QUESTION 13

Consider a regression model where the estimation results give the p-value of a coefficient as
“0.701”. At what levels of significance is this coefficient statistically significantly different from
zero?

(a) 1%.

(b) 10% but not 5% or 1%.

(c) Not at any conventional levels.

(d) 10% and 5% but not 1%.

QUESTION 14

Consider the regression model

CEOSAL = 2.613 + 0.45 * sales – 0.04 * sales2

Note that salesl2 = sales*sales. What is the effect of an increase from 2 to 3 in sales on
CEOSAL?

(a) 0.30

(b) 0.45

(c) 0.41

(c) 0.29

QUESTION 15

Consider the regression model

CEOSAL = 2.613 + 0.45 * sales – 0.04 * sales2

At what point of sales is CEOSAL maximized?

(a) 11.25

(b) 0.009

(c) 5.625

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 6 of 10

Page 6 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

(d) 0.018

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 7 of 10

Page 7 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

QUESTION 16

Which of the following is true of dummy/binary variables?

(a) A dummy variable always takes a value less than 1.

(b) A dummy variable always takes a value higher than 1.

(c) A dummy variable takes a value of 0 or 1.

(d) A dummy variable takes a value of 1 or 10.

QUESTION 17

A regression model has dummy variables for female and married:

log ( wage )=1.9+0.2∗married−0.7∗female−0.2∗female∗married
What is the predicted log(wage) for somebody who is female and married?

(a) 1.2

(b) 1.4

(c) 2.1

(d) 1.6

QUESTION 18

Consider the following regression equation: Yt = β0 + β1X1+…+ βk Xk + ut. In which of the

following cases is the dependent variable binary?

(a) The variable Yt indicates the gross domestic product of a country

(b) The variable Yt indicates whether an adult is employed

(c) The variable Yt indicates household consumption expenditure

(d) The variable Yt indicates the number of children in a family

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 8 of 10

Page 8 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

QUESTION 19

In the presence of heteroscedasticity, the usual OLS estimates of:

(a) standard errors are valid, whereas the t statistics are invalid.

(b) t statistics are valid, but the standard errors are invalid.

(c) standard errors and t statistics are both valid.

(d) standard errors and t statistics are both invalid.

QUESTION 20

Consider the model: yt = α0 + α1t1 + ut. What is the parameter α?

(a) the exponential time trend

(b) the lagged time trend

(c) the α time trend

(d) the linear time trend

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 9 of 10

Page 9 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

SECTION B

This section comprises three short-answer questions related to the STATA output shown
below. Students must answer all three questions in the script book provided. This section is
worth a total of 10 marks.

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 10 of 10

Page 10 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

QUESTION 1

In the STATA output above, lwage is the log of monthly earnings, educ is education in years,
exper is experience in years and educexper = educ*exper, the interaction of education and
experience. What is the predicted effect of 1 more year of education on wages when
educ=10 and exper=10? What is the predicted effect of 1 more year of education on wages
when educ=20 and exper=20?

(6 marks)

QUESTION 2

Using the STATA output above, interpret the meaning of the R-squared.

(2 marks)

QUESTION 3

Using the STATA output above, test the hypothesis that the coefficient on educ is different
from zero at the 1% level. Conduct a one-sided test. Use the critical value approach.

(2 marks)

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 11 of 10

Page 11 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

FORMULA SHEET

Critical values for the standard normal distribution (z)

Confidence Level of Two–Sided One-Sided, One-Sided,

level Significance Critical Value Upper-Tail Lower-Tail
(1-α) (α) cα/2 Critical Value Critical Value
cα -cα
90% 10% 1.645 1.28 -1.28
95% 5% 1.96 1.645 -1.645
99% 1% 2.58 2.33 -2.33

Formula for a t-statistic

estimate−hypothesised value
t=
standard error

Formula for a (1-α)% confidence interval

C I 1−α =( β−c
^ α / 2∗se ( ^β ) , ^β+ cα /2∗se ( ^β ) )

Logarithmic/Quadratic/Interaction specifications

For the model log ( y )= β^ 0 + ^β 1 x 1 + ^β 2 x 2, the exact effect of a change in explanatory variable x2
is:
[
% ∆ ^y =100 exp ( ^β 2 ∆ x 2 )−1 ]

For a quadratic specification of the form:

y=β 0 + β 1 x + β 2 x 2 +u

The turning point (maximum/minimum) is given by:

x ¿=| ^β1 /(2 β^ 2 )|

The approximation of the marginal effect of x on y is given by:

∆ ^y ^
≈ β 1+ 2 β^ 2 x
∆x

For a interaction specification of the form:

y=β 0 + β 1 x 1 + β 2 x 1∗x 2 +u
ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 12 of 10

Page 12 of 26

Downloaded by Quy?nh ???ng ([email protected])

 lOMoARcPSD|2960605

∆ ^y ^ ^
The approximation of the marginal effect of x1 on y is given by: ≈ β +β x
∆ x1 1 2 2

ECON1066 - Final Exam (Melb) Semester 1 Practice, 2018. Page 13 of 10

Page 13 of 26

Downloaded by Quy?nh ???ng ([email protected])