Scribd
Upload
2 views

Final Report Questions

The document is the final report for the EC2303 Foundations for Econometrics course at the National University of Singapore, containing instructions for students regarding the assessment format and submission guidelines. It includes six questions covering topics such as joint probability mass functions, marginal distributions, expectations, variances, and linear regression analysis. Students are required to show all derivations and submit their answers by a specified deadline.

Uploaded by

Habib Furqony
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2 views

Final Report Questions

The document is the final report for the EC2303 Foundations for Econometrics course at the National University of Singapore, containing instructions for students regarding the assessment format and submission guidelines. It includes six questions covering topics such as joint probability mass functions, marginal distributions, expectations, variances, and linear regression analysis. Students are required to show all derivations and submit their answers by a specified deadline.

Uploaded by

Habib Furqony
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

NATIONAL UNIVERSITY OF SINGAPORE

EC2303 Foundations for Econometrics


Final Report
Semester 2, AY 2020/2021

INSTRUCTIONS TO STUDENTS

1. The assessment contains six (6) questions and comprises four (4) printed pages
including this cover page. [Important] For the questions 1, 2, 3, please write your
derivations and answers in fractional numbers. For the question 4, you can use
decimal numbers up to 3 or 4 decimal places.
2. Answer sheets (PDF format preferred) should be submitted through: (1) LumiNUS
folder: Files\Report\Midterm report submission, and (2) Email to your tutor: you
can find your tutor’s email contact in the Files\tutorial folder. All submissions
should be done by 12:00pm (noon), 14th April. Note that late submissions will not
be considered for any reason. Please name your answer sheet as
EC2303_studentID_name.pdf for the online submission. For the email submissions,
please use the title “[EC2303] student ID, name”.
3. Your answers should include all the derivations (it means you should show your
work as much as possible). Correct answers without key derivations will not be
given full credits.
4. Throughout this evaluation, you may or may not need the standard normal table
in the Files\Lecture Slides folder. You can also refer to the lecture notes, slides
and other books. However, you are not allowed to discuss with other students. F
credit will be given for any evidence of collaboration (judged by the graders,
tutors and me).
5. If you need any clarification, please email me ([email protected]) between 12:30-
1:30pm, on 13th April. Or your inquiry is not guaranteed to be answered.

Good Luck!

1
1. (Total 8 points) Consider a discrete pair of random variables (X, Y ) with
joint PMF
1
fX,Y (x, y) = (x + y) for x = 0, 1 and y = 0, 1, 2, 3
16
= 0 otherwise.

(a) (1 point) What is the marginal PMF of Y , fY (y)?


(b) (1 point) What are E(Y ) and V ar(Y )?
(c) (1 point) What are E(XY ) and V ar(XY )?
(d) (1 point) What is Cov(X, Y )?
(e) (1 point) What is the conditional PMF of X given Y , i.e., fX|Y (x|y)?
(f) (1 point) What is the conditional CDF of X given Y , i.e., FX|Y (x|y)?
(g) (1 point) What are the conditional expectation and conditional variance
of X given Y = 3, i.e., E(X|Y = 3) and V ar(X|Y = 3)?
(h) (1 point) Derive the CDF of X 2 Y .

2. (Total 7 points) Consider a continuous pair of random variables (X, Y ) with


joint PDF
2
fX,Y (x, y) = (2x + 3y) for 0 ≤ x ≤ 1 and 0 ≤ y ≤ 1
5
= 0 otherwise.

(a) (1 point) What is the marginal PMF of Y , fY (y)?


(b) (1 point) What are E(Y ) and V ar(Y )?
(c) (1 point) What are E(XY ) and V ar(XY )?
(d) (1 point) What is Cov(X, Y )?
(e) (1 point) What is the conditional PDF of X given Y , i.e., fX|Y (x|y)?
(f) (1 point) What is the conditional CDF of X given Y , i.e., FX|Y (x|y)?
(g) (1 point) Calculate E(X|Y = 21 ) and V ar(X|Y = 12 ).

2
3. (Total 6 points) Let (X, Y ) be a pair of continuous random variables which
take values on [0, 1] × [0, 1]. For some constant θ ∈ [0, 1], the joint cdf of X
and Y is provided by
xy
FXY (x, y) = for 0 ≤ x ≤ 1 and 0 ≤ y ≤ 1.
1 − θ(1 − x)(1 − y)
1
(a) (1 point) What is P (X ≤ 2
and Y ≤ 13 )?
1
(b) (1 point) What is P (X > 2
and Y < 13 )?
(c) (2 points) What do you know about the distribution of X? And what
do you know about the distribution of Y ? (Hint: find the CDF of X
and the CDF of Y ).
(d) (2 points) Suppose that θ = 0. Calculate Corr(X, Y ).

4. (Total 6 points) Suppose that X1 ,X2 ,... are iid random variables, each of
which is equal to zero with probability 23 , and equal to one with probability
1
3
.

(a) (2 points) Let µ = E(Xi ), and let Zn denote the quantity


n
1 X
Zn = √ (Xi − µ).
n i=1
√
When n = 2, what is the probability that Zn ≤ 0.924? (Hint: 2≈
1.4142)
(b) (4 points) How will your answer to (a) change as n → ∞? (In other
words, what happens to P (Zn ≤ 0.924) as n → ∞?)

3
5. (Total 4 points) Suppose you have iid pairs of sample (X1 , Y1 ), (X2 , Y2 ),...,
(Xn , Yn ). You wish to estimate a function M (t1 , t2 ) defined by

M (t1 , t2 ) = E(et1 X+t2 Y )

where e is the natural number. Note that M (t1 , t2 ) is a bivariate function


of t1 and t2 after taking the expectation. Provide a consistent estimator of
M (t1 , t2 ) and show the consistency.

6. (Total 9 points) Suppose we estimate a linear regression equation Yi = β0 +


β1 Xi + ui by OLS.

(a) (3 points) Show that n1 ni=1 ûi = 0, where the ûi ’s are the regression
P
residuals.
(b) (3 points) Suppose we regress Xi on ûi by OLS, including a constant
term in the regression. Show that the estimated coefficient on ûi is equal
to zero.
(c) (3 points) Suppose we regress Yi on the predicted value Ŷi by OLS,
including a constant term in the regression. Show that the estimated
coefficient on Ŷi is equal to one.

(Hint: the regression residual is defined by ûi = Yi − Ŷi where the predicted
value Ŷi = β̂0 + β̂1 Xi . Here, β̂0 and β̂1 are the OLS estimators.)

You might also like