ECONOMETRICS - FINAL EXAM

(GADE)

May 08, 2023 – 12:00 AM

Family name: Name:

DNI/ID: Instructor:

Question 1 A B C Blank
Question 2 A B C Blank
Question 3 A B C Blank
Question 4 A B C Blank
Question 5 A B C Blank
Question 6 A B C Blank
Question 7 A B C Blank
Question 8 A B C Blank
Question 9 A B C Blank
Question 10 A B C Blank
Question 11 A B C Blank
Question 12 A B C Blank
Question 13 A B C Blank
Question 14 A B C Blank
Question 15 A B C Blank
Question 16 A B C Blank
Question 17 A B C Blank
Question 18 A B C Blank
Question 19 A B C Blank
Question 20 A B C Blank

Correct Incorrect Blank Final grade

-1-
 INSTRUCTIONS

This exam includes 20 multiple choice questions.

Your answers must be marked on the answer sheet that you will find in the first page.
If you want to leave any question unanswered, choose the "Blank" option. This answer
sheet is the only part of this exam that will be graded.

A correct answer adds 2 points to the final grade while an incorrect one subtracts 1
point. A blank answer does not add or subtract. The final grade is the number of points
divided by 4.

Make sure that you checked your options, including “Blank”. Do not unclip the sheets.
Use the blank space in the following pages to write notes or to do arithmetic
calculations.

YOU HAVE ONE HOUR TO ANSWER THIS TEST

REMINDER
YOU ARE NOT ALLOWED TO USE DEVICES WITH
CONNECTIVITY TO THE INTERNET, INCLUDING MOBILE
PHONES, TABLETS, SMARTWATCHES OR MP3/4 PLAYERS

-2-
Question 1. In the model 𝑌𝑌� = 𝛽𝛽̂1 + 𝛽𝛽̂2 𝑋𝑋2 + 𝛽𝛽̂3 𝑋𝑋3, which of the following statements is NOT
ALWAYS TRUE?

A) If the sample average of Y , X 2 y X 3 are equal to zero, then 𝛽𝛽̂1 = 0.

B) The Sum of Squared Residual is equal or greater than zero, 𝑆𝑆𝑆𝑆𝑆𝑆 ≥ 0.
C) If we estimate the model 𝑌𝑌 = 𝛼𝛼1 + 𝛼𝛼2 𝑋𝑋2 + 𝑢𝑢, The OLS estimate of 𝛼𝛼2 is equal to 𝛽𝛽̂2 .

Question 2. To obtain a confidence interval for 𝛽𝛽2 in the model 𝑌𝑌 = 𝛽𝛽1 + 𝛽𝛽2 𝑋𝑋2 + 𝑈𝑈, we need to
estimate 𝛽𝛽̂2 , the standard error of 𝛽𝛽̂2 and set the level of confidence (1 − 𝛼𝛼). Then, which of the
following statements is FALSE. The width (length) of the interval will increase with the …

A) size of 𝛽𝛽̂2 .
B) standard error of 𝛽𝛽̂2 .
C) level of confidence.

Question 3. Regarding the Variance Inflation Factor (VIFj) for the slope coefficient j in the
regression 𝑌𝑌 = 𝛽𝛽1 + 𝛽𝛽2 𝑋𝑋2 + 𝛽𝛽3 𝑋𝑋3 + 𝑈𝑈, which of the following statements is TRUE.

A) VIF3 = 1 reveals a MULTICOLLINIARITY “problem” in this model.

B) VIF3 = 1 reveals a HETEROSKEDASTICITY “problem” in this model.
C) None of the above.

Question 4. The existence of heteroskedasticity is problematic since (which of the following

statements is TRUE):

A) The OLS estimator becomes biased.

B) The usual estimate of the variance of the OLS estimator will be biased.
C) A multicollinearity-robust standard error estimator should be calculated.

Question 5. Influential observations are defined as those that have…

A) a large impact on the OLS estimates.

B) a high leverage.
C) a large OLS residual.

-3-
Questions 6 to 9 correspond to the following statement. Consider the following model:

β1 β 2 X i 2 + β3 X i 3 + U i , i =
Yi =+ 1,..., 44

And the next tables that show the OLS estimates of parameters and the estimated variance
and covariance matrix of the OLS estimates.

 βˆ1   −27.05986  32.057666 . .

    ̂
 βˆ2  =  1.498472  𝑣𝑣𝑎𝑎� 𝑟𝑟( 𝛽𝛽 ) = �−0.1290654 0.00979752 . �
 ˆ   1.27375  −0.1552826 −0.0069964 0.0070221
 β 3   
𝜎𝜎�𝑢𝑢2 = 5.9734; 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝑆𝑆𝑆𝑆𝑆𝑆 𝑜𝑜𝑜𝑜 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 = 𝑆𝑆𝑆𝑆𝑆𝑆 = 17565.3795

Note: Use all available decimals to answer question 6 to 9

Question 6. which of the following statements is TRUE:

A) Total Sum of Squares (SST)= 17560.4061.
B) R2 = 0.986.
C) Sum of Squared Residuals (SSR) = 444.91.

Question 7. The LARGEST associated p-values for the individual t test of the null hypothesis
H 0 : βi = 0 against the alternative H1 : βi ≠ 0 is:

A) for β 3 .
B) for β 2 .
C) for β1 .

Question 8. Critical values are reported as: Pr[t(41) ≤ 2.4208] = 0.99; Pr[t(41) ≤ 1.6829] = 0.95
and Pr[t(41) ≤ 1.3025] = 0.90. According to the information provided above, the test of the
null hypothesis 𝐻𝐻0 : 𝛽𝛽2 = 1 against the alternative 𝐻𝐻1 : 𝛽𝛽2 > 1 indicates that:

A) The null is rejected at both 1% and 5% level of significance.

B) The numerical value of the statistic (t*) for this hypothesis is 15.14.

C) We fail to reject the null at both 5% and 10% level of significance.

-4-
Question 9. Critical values are reported as Pr[F(2,41) ≥ 3.2257] = 0.05. According to the
information provided above, the null hypothesis that none of the explanatory has an effect on
Y (all the slope parameters are zero) …

A) is rejected at 5% level of significance because the F-statistic for this hypothesis is =

1470.30.

B) is NOT rejected at 5% level of significance because the F-statistic for this hypothesis is
= 1.64.

C) is rejected at 5% level of significance because the F-statistic for this hypothesis is =

104.34.

Question 10. Birth weight (bwght) is a measure of infant health. Infants with low birth weight
are often more vulnerable to disease. One problem of interest is the relationship between
maternal smoking and the child’s birth weight. To analyze that, a researcher estimates the
model [M1]

𝑏𝑏𝑏𝑏𝑏𝑏ℎ𝑡𝑡𝑖𝑖 = 𝛽𝛽1 + 𝛽𝛽2 𝑐𝑐𝑐𝑐𝑐𝑐𝑠𝑠𝑖𝑖 + 𝑒𝑒𝑖𝑖 .

The researcher thinks that factors other than number of cigarettes smoked (𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐) that affect
birth weight are likely to be correlated with smoking, and therefore those factors should be
considered. For example, higher income (faminc) generally results in access to better prenatal
care, as well as better nutrition for the mother. To analyze this effect a new model [M2] is
estimated:

𝑏𝑏𝑏𝑏𝑏𝑏ℎ𝑡𝑡𝑖𝑖 = 𝛼𝛼1 + 𝛼𝛼2 𝑐𝑐𝑐𝑐𝑐𝑐𝑠𝑠𝑖𝑖 + 𝛼𝛼3 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑛𝑛𝑐𝑐𝑖𝑖 + 𝑒𝑒𝑖𝑖 .

According to this, which of the following statements is TRUE?

A) The OLS estimate of 𝛼𝛼2 in [M2] is always equal to the OLS estimate of 𝛽𝛽2 in [M1]
regardless the value of the corr(cigs, faminc).

B) The OLS estimate of 𝛼𝛼2 in [M2] will be lower in absolute value than the OLS estimate
of 𝛽𝛽2 in [M1] if corr(cigs, faminc) < 0.

C) The OLS estimate of 𝛼𝛼2 in [M2] will be lower in absolute value than the OLS estimate
of 𝛽𝛽2 in [M1] if corr(cigs, faminc) > 0.

-5-
Questions 11 to 13 correspond to the following statement. Using a sample of 10 workers of
three different sectors (agriculture, services, and industry), a regression model [W1] was
estimated to explain the behaviour of wages.

ln(𝑤𝑤𝑖𝑖 ) = 2.11 − 0.56𝐴𝐴𝑖𝑖 − 0.45𝑆𝑆𝑖𝑖 + 0.12 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖 + 0.08𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖 + 𝑢𝑢�𝑖𝑖 [𝑾𝑾𝑾𝑾]

Where, w is the monthly salary (euros), A is a dummy variable that takes a value of 1 if the
worker belongs to the agriculture sector and zero otherwise. S is a dummy variable that takes
a value of 1 if the worker belongs to the services sector and zero otherwise, educ is years of
education and exp is years of professional experience.

Question 11. According to the model W1, which of the following statements is TRUE?
A) Ceteris paribus, the monthly salary for a worker who belongs to the agriculture sector
is approximately 0.56% less than the rest of workers.

B) Ceteris paribus, the monthly salary for a worker who belongs to the agriculture sector
is approximately 56.00 % less than a worker that belongs to the industrial sector.

C) Ceteris paribus, the monthly salary for a worker who belongs to the services sector is
approximately 0.45% less than a worker that belongs to the industrial sector.

Question 12. If the estimate of the standard deviation of the effect of education on wage (𝛽𝛽𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 )
is �(𝛽𝛽̂𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 ) = 0.10 and Pr[-2.57≤ t(5) ≤ 2.57]=0.95, which of the following statements is
𝑠𝑠𝑠𝑠
TRUE?
A) A 95% confidence interval for 𝛽𝛽𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 is [0.02, 0.22]
B) There is not enough information to compute a 95% confidence Interval for 𝛽𝛽𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 .
C) The null hypothesis 𝐻𝐻0 : 𝛽𝛽𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 = 0 against 𝐻𝐻1 : 𝛽𝛽𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 ≠ 0 cannot be rejected at 5% level
of significance.

Question 13. A beginner student in Econometrics estimates three different equations to test for
heteroscedasticity in W1. Could you help this student to choose the right equation? (Choose
the right one).
A) ln(𝑤𝑤𝑖𝑖 ) = 𝛾𝛾0 + 𝛾𝛾1 ln�
(𝑤𝑤𝚤𝚤 ) + 𝛾𝛾2 ln�
(𝑤𝑤𝚤𝚤 )2 + 𝑒𝑒𝑖𝑖 , where ln�
(𝑤𝑤𝚤𝚤 ) are the fitted values in W1.

B) 𝑢𝑢�𝑖𝑖 2 = 𝛼𝛼0 + 𝛼𝛼1 𝐴𝐴𝑖𝑖 + 𝛼𝛼2 𝑆𝑆𝑖𝑖 + 𝛼𝛼3 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖 + 𝛼𝛼4 𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖 + 𝛼𝛼5 𝑒𝑒𝑑𝑑𝑑𝑑𝑑𝑑 2 𝑖𝑖 + 𝛼𝛼6 𝑒𝑒𝑥𝑥𝑥𝑥𝑥𝑥𝑥𝑥 2 𝑖𝑖 + 𝛼𝛼7 𝐴𝐴𝑖𝑖 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖 +

𝛼𝛼8 𝑆𝑆𝑖𝑖 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖 + 𝛼𝛼9 𝐴𝐴𝑖𝑖 𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖 + 𝛼𝛼10 𝑆𝑆𝑖𝑖 𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖 + 𝛼𝛼11 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖 𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖 + α12 Ai Si + ξi , where 𝑢𝑢�𝑖𝑖 are the

residual in W1.

C) 𝑢𝑢�𝑖𝑖 = 𝛿𝛿0 + 𝛿𝛿1 ln�

(𝑤𝑤𝚤𝚤 ) + 𝛿𝛿2 ln�
(𝑤𝑤𝚤𝚤 )2 + 𝑒𝑒𝑖𝑖 , where 𝑢𝑢�𝑖𝑖 are the residual and ln�
(𝑤𝑤𝚤𝚤 ) the fitted

values in W1.

-6-
Questions 14 and 15 refer to the following statement. The manager of Ben&Jerry’s wants to
analyze if sales of ice creams (Y) exhibit seasonality (for example, sales in the first quarter are
lower than in the other three quarters). They are going to use quarterly data (4 observation
per year) from the first quarter of 2000 to the third of 2022, both included. A model has been
specified including 4 seasonal dummy variables as regressors, Dit’s, taking the value of 1 in the
i-th quarter (i = 1, 2, 3, 4) and 0 otherwise.

Question 14. If the manager decides to estimate the next model:

𝑌𝑌𝑡𝑡 = 𝛽𝛽0 + 𝛽𝛽1 𝐷𝐷1𝑡𝑡 + 𝛽𝛽2 𝐷𝐷2𝑡𝑡 + 𝛽𝛽3 𝐷𝐷3𝑡𝑡 + 𝛽𝛽4 𝐷𝐷4𝑡𝑡 + 𝑢𝑢𝑡𝑡

Which of the following statements is TRUE?

A) The manager might have introduced perfect collinearity in the model, and this is a
serious issue for applying OLS to forecast sales.

B) The manager might have introduced high collinearity (but not perfect) in the model,
and this is a serious issue for applying OLS to forecast sales.

C) None of the above

Question 15. If the manager decides to estimate the next model:

𝑌𝑌𝑡𝑡 = 𝛼𝛼0 + 𝛼𝛼1 𝐷𝐷1𝑡𝑡 + 𝛼𝛼2 𝐷𝐷2𝑡𝑡 + 𝛼𝛼3 𝐷𝐷3𝑡𝑡 + 𝑢𝑢𝑡𝑡

To test whether or not sales exhibit seasonality, the manager should estate one of the next
hypotheses. (Choose the right one):

A) 𝐻𝐻0 : 𝛼𝛼0 = 0

B) 𝐻𝐻0 : 𝛼𝛼0 = 𝛼𝛼1 = 𝛼𝛼2 = 𝛼𝛼3

C) 𝐻𝐻0 : 𝛼𝛼1 = 𝛼𝛼2 = 𝛼𝛼3 = 0

-7-
Questions 16 to 17 refer to the following statement. A broker wants to analyze if the returns
of the Bank of Santander (SAN) depend on the returns of the IBEX35 and estimates the next
model [M1]:

𝑆𝑆𝑆𝑆𝑁𝑁𝑡𝑡 = 0.609 + 0.78𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼35𝑡𝑡 + 𝑢𝑢�𝑡𝑡

𝑁𝑁 = 79, 𝑅𝑅 2 = 0.45, 𝑊𝑊𝑊𝑊 = 0.894 [𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 = 0.639]

Questions 16. The White statistic (WH) suggests that the error term could have … (Select the
True option):

A) Heteroskedasticity.

B) Homoskedasticity.
C) Homoskedasticity and multicollinearity.

Question 17. The estimate of the variance and covarince matrix or the errors (ui) of model
M1is:

𝟏𝟏𝟏𝟏 𝟎𝟎 ⋯ 𝟎𝟎
𝟎𝟎 𝟏𝟏𝟏𝟏 ⋯ 𝟎𝟎
𝑴𝑴𝑴𝑴𝑴𝑴[𝑼𝑼] = � �
⋮ ⋮ ⋱ ⋮
𝟎𝟎 𝟎𝟎 ⋯ 𝟏𝟏𝟏𝟏

Then, which of the following statements is TRUE?

A) The Gauss-Markov assumption 𝑉𝑉𝑉𝑉𝑉𝑉(𝑢𝑢𝑖𝑖 ) = 𝜎𝜎 2 , for i=1, …, 79, (AS 6) holds.

B) The Gauss-Markov assumption 𝐶𝐶𝐶𝐶𝐶𝐶�𝑢𝑢𝑖𝑖 , 𝑢𝑢𝑗𝑗 � = 0, (AS7) DOES NOT hold.
C) None of the above.

Questions 18. A master student found that there is a high correlation between measles (a highly
contagious, serious disease caused by a virus) cases and Marriage rate in the U.S., see Figure
MM1.
Figure MM1

-8-
From Figure MM1 (U.S. Marriage rate is the top one) you could estate that (select the true
statement)

A) Lower numbers of measles cases are reducing (causality) marriage rates in the U.S.

B) Lower rates of marriage are reducing (causality) the numbers of measles cases in
the U.S.

C) This might be an example of spurious relationship. i.e., high correlation between

these two variables exists by chance and there is no causality.

Question 19. In the model 𝑦𝑦𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1 𝑥𝑥𝑖𝑖 + 𝑢𝑢𝑖𝑖 , which of the following statements is TRUE?

A) It would not be possible to compute the OLS estimate of 𝛽𝛽1 if the sample 𝑣𝑣𝑣𝑣𝑣𝑣(𝑥𝑥𝑖𝑖 ) = 0.
B) The OLS estimate of 𝛽𝛽1 would be equal to 1 if the sample 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐(𝑦𝑦𝑖𝑖 , 𝑥𝑥𝑖𝑖 ) = 1.
C) The OLS estimate of 𝛽𝛽1 would be equal to the 𝑦𝑦� (y sample average) if the 𝑥𝑥̅ = 0.

Question 20. The assumption that the error terms in a regression model follow the normal
distribution is required for … (choose the TRUE statement)
A) unbiasedness of the OLS estimator.
B) efficiency (minimum variance) of the OLS estimator.
C) hypothesis testing and inference.

-9-
Calculations

- 10 -
 ECONOMETRICS - FINAL EXAM, 3rd YEAR (GADE)

May 08, 2023 – 12:00 AM

Family name: Name:

DNI/ID: Instructor:
Mobile: E-mail:

Question 1 A B C Blank
Question 2 A B C Blank
Question 3 A B C Blank
Question 4 A B C Blank
Question 5 A B C Blank
Question 6 A B C Blank
Question 7 A B C Blank
Question 8 A B C Blank
Question 9 A B C Blank
Question 10 A B C Blank
Question 11 A B C Blank
Question 12 A B C Blank
Question 13 A B C Blank
Question 14 A B C Blank
Question 15 A B C Blank
Question 16 A B C Blank
Question 17 A B C Blank
Question 18 A B C Blank
Question 19 A B C Blank
Question 20 A B C Blank

Correct Incorrect Blank Final grade

- 11 -