WILFRID LAURIER UNIVERSITY

WATERLOO | BRANTFORD | KITCHENER | TORONTO

MGT604 (Introductory Econometrics)

Instructor: Jean Eid

Practice Exam

Duration:

Aids Allowed: Non-programmable calculator

This is practice exam for Introductory Econometrics(MGT604). Please answer the questions as clearly and in as much detail as
possible.
This examination consists of 11 pages with 4 questions

Please bring any discrepancy to the attention of an invigilator. The number in parentheses at the start of each question is the
number of points the question is worth.

Answer the 4 questions.

75 University Avenue West, Waterloo, Ontario N2L3C5 LazaridisSchool.ca

T: (519) 884-0710
 Short Answer and Calculation Questions
Question 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Question 1
Interpretation and hypothesis testing of regression coefficients
Suppose you have a cross sectional data set on smoking with the following variables

Var Label

educ years of schooling

cigpric cig. price in cents per pack

white =1 if white

age in years

income annual income in thousands of $

cigs cigs. smoked per day

restaurn =1 if rest. smk. restrictions

lincome log(income)

agesq age^2

lcigpric log(cigprice)

(a) To start with, you issue these commands in

R Chunk: 1

> reg1 <-lm(cigs~cigpric+age+white+lincome, data=smoke)

> summary(reg1)

Question 1 continues Practice Exam Page 1 of 11

on next page (MGT604–Fall–2023)
 R Output for Chunk: 1

Call:
lm(formula = cigs ~ cigpric + age + white + lincome, data = smoke)

Residuals:
Min 1Q Median 3Q Max
-10.520 -9.112 -7.695 10.216 70.566

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 8.72317 6.65075 1.312 0.1900
cigpric -0.04264 0.10246 -0.416 0.6774
age -0.03074 0.02842 -1.082 0.2796
white 0.32279 1.48187 0.218 0.8276
lincome 1.26599 0.68062 1.860 0.0632

Residual standard error: 13.71 on 802 degrees of freedom

Multiple R-squared: 0.00615, Adjusted R-squared: 0.001193
F-statistic: 1.241 on 4 and 802 DF, p-value: 0.2921

Interpret the regression coefficients (saying slope and intercept will earn you no marks, saying negative or positive corre-
lation/association earns you zero marks)
(b) Set up the null and alternative hypotheses to test whether or not cigarette prices decrease the average daily consumption
of cigarettes.
(c) In plain English and in terms of the cigarette price coefficient, state the rejection rule of the hypothesis above. This rule
should be stated in general and before we run the regression.
(d) Test the hypothesis above at the level α = 5%. You can get the critical value for the test statistic from the figure
below. % on figures indicate probability of shaded area for that side only. When conducting the test, state the degrees of
freedom of the distribution

2.5% 2.5% 5% 5%

−∞ −1.96 1.96 ∞ −∞ −1.65 1.65 ∞

10% 10%

−∞ −1.28 1.28 ∞

(e) Set up the null and alternative hypotheses to test whether or not white people on average smoke more than others.
(f) In plain English and in terms of the white variable coefficient, state the rejection rule of the hypothesis above. This rule
should be stated in general and before we run the regression.
(g) Test the hypothesis above at the level α = 2.5%. You can get the critical value for the test statistic from the figure
below. % on figures indicate probability of shaded area for that side only. When conducting the test, state the degrees of

Question 1 continues Practice Exam Page 2 of 11

on next page (MGT604–Fall–2023)
 freedom of the distribution

2.5% 2.5% 5% 5%

−∞ −1.96 1.96 ∞ −∞ −1.65 1.65 ∞

10% 10%

−∞ −1.28 1.28 ∞

(h) place your test statistics in the figure below and draw its p-value.

−∞ ∞
(i) Test at the level of α = 5% whether or not on average people will smoke zero cigarettes if the price of cigarettes is zero
dollars.
(j) Suppose you have in addition to the regression above, you have the following regressions in . In addition, Suppose the
critical value is 2.6160051.
R Chunk: 2

> reg2 <- lm(cigs~cigpric, data=smoke)

> summary(reg2)

MGT604–Fall–2023 Practice Exam Page 3 of 11

 R Output for Chunk: 2

Call:
lm(formula = cigs ~ cigpric, data = smoke)

Residuals:
Min 1Q Median 3Q Max
-9.224 -8.678 -8.575 11.082 71.332

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 10.67457 6.17296 1.729 0.0841
cigpric -0.03297 0.10206 -0.323 0.7467

Residual standard error: 13.73 on 805 degrees of freedom

Multiple R-squared: 0.0001296, Adjusted R-squared: -0.001112
F-statistic: 0.1044 on 1 and 805 DF, p-value: 0.7467

Test that only cigpric affect the consumption of cigs and the rest of the variables do not matter
Question 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Question 2
Interpretation of regression coefficients
(a) Suppose we use the same data set as the one from the previous question and have the following commands in
R Chunk: 3

> smoke <- smoke%>%

> mutate(smoke=cigs>0, smokewhite=smoke*white, educwhite=educ*white)
> temp <- mysummary(smoke)%>%
> set_background_color(row=which(c(1:nrow(.))%%2==0), col=everywhere, "grey")%>%
> set_background_color(row=which(!c(1:nrow(.))%%2==0)[-1], col=everywhere, "grey90")
> cat(to_latex(temp, tabular_only=T), "\n")

Variable Obs Mean Std. dev. Min Max

educ 807 12.47088 3.057161 6 18
cigpric 807 60.30041 4.738469 44.004 70.129
white 807 0.8785626 0.3268375 0 1
age 807 41.23792 17.02729 17 88
income 807 19.30483 9.142958 0.5 30
cigs 807 8.686493 13.72152 0 80
restaurn 807 0.2465923 0.4312946 0 1
lincome 807 2.779561 0.7126959 -0.6931472 3.401197
agesq 807 1990.135 1577.166 289 7744
lcigpric 807 4.096032 0.08291944 3.784281 4.250336
smoke 807 0.3841388 0.4866926 0 1
smokewhite 807 0.33829 0.4734212 0 1
educwhite 807 10.95353 4.972069 0 18
i. What is the proportion of white people in the sample. Given this proportion, how representative is this sample of the
Canadian population? (explain your answer)
ii. Approximately how many restaurants have smoking restrictions?
iii. How many white people smoke in the sample

Question 2 continues Practice Exam Page 4 of 11

on next page (MGT604–Fall–2023)
 iv. Out of all the white people, what is the proportion that smoke
(b) We now issue the following commands in and put the results in the following table. What do you attribute the change
in the coefficients from the first column to each of the corresponding coefficient in the rest.
R Chunk: 4

> reg_cigs1 <- lm(cigs~ lincome+lcigpric+educ+white+educwhite, data=smoke)

> reg_cigs2 <- lm(cigs~ lincome, data=smoke)
> reg_cigs3 <- lm(cigs~ lcigpric, data=smoke)
> reg_cigs4 <- lm(cigs~ educ, data=smoke)
> reg_cigs5 <- lm(cigs~ white, data=smoke)
> reg_cigs6 <- lm(cigs~ educwhite, data=smoke)

R Chunk: 5

> temp <- myht(list(reg_cigs1, reg_cigs2 , reg_cigs3, reg_cigs4, reg_cigs5, reg_cigs6),

,→ names=c("cigs_1", "cigs_2", "cigs_3", "cigs_4", "cigs_5", "cigs_6"), latex=F,
,→ stars=NULL)
> temp[1,1] <- "dep. var: \\mathbox{\\texttt{cigs}}"
> temp <- set_escape_contents(temp, row=1, col=1, value=F)
> cat(to_latex(temp, tabular_only=T), "\n")

dep. var: cigs cigs_1 cigs_2 cigs_3 cigs_4 cigs_5 cigs_6

(Intercept) 19.955 5.155 18.592 11.412 8.408 9.281
(24.773) (1.943) (23.892) (2.029) (1.387) (1.170)
lincome 1.769 1.271
(0.715) (0.677)
lcigpric -3.056 -2.418
(5.850) (5.832)
educ -0.322 -0.219
(0.437) (0.158)
white 0.723 0.317
(6.030) (1.480)
educwhite -0.027 -0.054
(0.468) (0.097)
N 807 807 807 807 807 807
2
R 0.010 0.004 0.000 0.002 0.000 0.000
2
adj. R 0.004 0.003 -0.001 0.001 -0.001 -0.001
i. Carefully interpret the coefficients on lincome, lcigpric, in the first column
ii. What is the marginal effect of education on the number of cigarettes smoked for a white person
iii. What is the marginal effect of education on the number of cigarettes for a non-white person
iv. What does the coefficient on white alone in the first column represents
v. What does the constant in the first column represents
(c) We now issue the following commands in and output the regression table. and put the results in the following table.

Question 2 continues Practice Exam Page 5 of 11

on next page (MGT604–Fall–2023)
 R Chunk: 6

> reg_smoke1 <- lm(smoke~lincome+lcigpric+educ+white+educwhite, data=smoke)

> reg_smoke2 <- lm(smoke~lincome, data=smoke)
> reg_smoke3 <- lm(smoke~lcigpric, data=smoke)
> reg_smoke4 <- lm(smoke~educ, data=smoke)
> reg_smoke5 <- lm(smoke~white, data=smoke)

R Chunk: 7

> temp <- myht(list(reg_smoke1, reg_smoke2 , reg_smoke3, reg_smoke4, reg_smoke5),

,→ names=c("smoke_1", "smoke_2", "smoke_3", "smoke_4", "smoke_5"), latex=F, stars=NULL)
> temp[1,1] <- "dep. var: \\mathbox{\\texttt{smoke}}"
> temp <- set_escape_contents(temp, row=1, col=1, value=F)
> cat(to_latex(temp, tabular_only=T), "\n")

dep. var: smoke smoke_1 smoke_2 smoke_3 smoke_4 smoke_5

(Intercept) 1.271 0.370 1.053 0.627 0.378
(0.875) (0.069) (0.847) (0.072) (0.049)
lincome 0.036 0.005
(0.025) (0.024)
lcigpric -0.161 -0.163
(0.207) (0.207)
educ -0.027 -0.020
(0.015) (0.006)
white -0.061 0.007
(0.213) (0.052)
educwhite 0.005
(0.017)
N 807 807 807 807 807
2
R 0.018 0.000 0.001 0.015 0.000
2
adj. R 0.012 -0.001 -0.000 0.014 -0.001
i. What do you attribute the change in the coefficients from the first column to each of the corresponding coefficient
in the rest.
ii. Carefully interpret the coefficients on lincome, lcigpric, in the first column
iii. What is the marginal effect of education on the dependent variable for a white person
iv. What is the marginal effect of education on the number of cigarettes for a non-white person
v. What does the coefficient on white alone in the first column represents
vi. What does the constant in the first column represents
Question 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Question 3
Scaling variables
Suppose we issue the following commands in . The information needed to finish this question (regression tables and summary
statistics) is given in the previous question

Question 2 continues Practice Exam Page 6 of 11

on next page (MGT604–Fall–2023)
 R Chunk: 8

> smoke <- smoke%>%

> mutate(income2=income*1000, lincome2=log(income2), packs=cigs/20, other=1-white,
,→ educ1=educ-12)
> my.summary.lm(summary(lm(cigs~lincome+lcigpric+educ+white , data=smoke)))

R Output for Chunk: 8

Call:
lm(formula = cigs ~ lincome + lcigpric + educ + white, data = smoke)

Residuals:
Min 1Q Median 3Q Max
-12.336 -9.064 -7.624 9.254 72.114

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 20.3060 24.0093 0.846 0.3979
lincome 1.7680 0.7144 2.475 0.0135
lcigpric -3.0686 5.8416 -0.525 0.5995
educ -0.3450 0.1661 -2.077 0.0381
white 0.3843 1.4784 0.260 0.7950

Residual standard error: 13.69 on 802 degrees of freedom

Multiple R-squared: 0.01014, Adjusted R-squared: 0.005206
F-statistic: 2.054 on 4 and 802 DF, p-value: 0.085

(a) Suppose we run a regression of

cigsi = γo + γ1 lincome2i + γ2 lcigpric + γ3 educ1 + γ4 other + i

bo , γ
i. Calculate the value of γ b1 , γ
b2 , γ
b3 , γ
b4
ii. What happens to the fitted values and residuals (explain)
Question 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Question 4
Consider the following regression output

Question 2 continues Practice Exam Page 7 of 11

on next page (MGT604–Fall–2023)
 dependent: S Coefficient Std. err. t-stat p>|t| [95% conf. interval]

(Intercept) Calculate 0.514 9.664 0.000 Calculate 5.978

SM 0.050 Calculate 1.268 0.205 -0.027 Calculate

SF 0.107 0.031 Calculate 0.001 Calculate 0.168

ASVAB02 Calculate 0.013 5.886 0.000 Calculate Calculate

Calculate Calculate Calculate 2.126 0.034 0.003 0.068

Calculate 0.026 0.015 1.736 0.083 Calculate Calculate

R2 Calculate SSR 2021.84

adj. R2 0.363 ESS 1183.15

sigma Calculate

F-stat Calculate

N 540
(a) Calculate and complete the above table/calculate all values where you see the "Calculate" word. Assume the t534,0.025 =
1.9645.
Provide your calculations below.

Question 2 continues Practice Exam Page 8 of 11

on next page (MGT604–Fall–2023)