0% found this document useful (0 votes)
13 views3 pages

E Ec605d22 23

The document is an exam paper from Maynooth University's Department of Economics for the EC605D course, consisting of four questions that cover topics such as joint probability distributions, OLS estimators, variance-covariance matrices, and regression analysis related to crime rates. Each question has specific parts that require calculations and explanations, with varying weights assigned to each question. The exam duration is 1 hour and 45 minutes, and students are instructed to turn off communication devices during the exam.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
13 views3 pages

E Ec605d22 23

The document is an exam paper from Maynooth University's Department of Economics for the EC605D course, consisting of four questions that cover topics such as joint probability distributions, OLS estimators, variance-covariance matrices, and regression analysis related to crime rates. Each question has specific parts that require calculations and explanations, with varying weights assigned to each question. The exam duration is 1 hour and 45 minutes, and students are instructed to turn off communication devices during the exam.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

MAYNOOTH UNIVERSITY

DEPT OF ECONOMICS
EC605D Final Exam 2022/23

Answer all questions. You must explain your answer in order to get marks.

Q1 is worth 25%. Question 2 is worth 15%. Question 3 is worth 15%. Question 4 is


worth 45%
Put your phone and any other communications device on the desk and turn them off.
Time: 1 hour 45 minutes.

1. Students take two exams, Science(S) and Maths(M). Assume that they can only get 3
possible grades on the exam which are associated with points: Fail (30 points) Pass
(50 points) and Distinction(100 points) . The joint probability distribution of M and S
is shown below

Points in Science (S)


30 50 100
Points 30 0.05 0.24 0.05
in
50 0.1 0.18 0.21
Maths
(M) 100 0.02 0.05 0.1

Be sure to show all your calculations

a) Find E(S), E(M) and Std(S) . From your results, which subject do you think
students might find it easier to get higher marks? Explain.
[ 40 points]
b) Find E(S|M=30) and E(S|M=100) . Compare these results, and explain what they
mean. You can assume that you are explaining it to someone who was not in class
the day conditional expectations were covered. [30 points]
c) Are S and M independent? Explain your answer, no marks for unexplained answer
. [30 points]

2 Using the assumptions below ( on page 2),


−1
a) Show that the Variance-Covariance matrix of the OLS estimator b =σ 2 ( X ' X ) i.e.
−1
show that E (b−E ( b ) ) ¿ = σ 2 ( X ' X ) . Note: The formula for the OLS estimator, b,
−1
is ( X ' X ) ( X ' y ) You can assume that the OLS estimator is unbiased so E(b) =β.
[70 points]
b) Suppose you had estimated the Var-Cov data using a sample of data. How would you
find the standard errors for the estimators from this matrix. [30 points]
See assumptions below.

Page 1 of 3
Assumptions.
1. E(y) =Xβ, y =Xβ+ε→ E(ε)=0
2. V(y) = V(ε )=σ2I :V(y) is defined as E(y-X β)(y-X β)’ = E(ε ε’)
3. X is non-stochastic.
4. Rank(X) =k

3 Suppose students who study French take both an oral exam and a written exam at the end
of the semester. Their total mark(T) is calculated as 0.6*R +0.4*W, where R is the mark for
the Oral exam and W is the mark for the written exam. suppose E(R)=80% and E(W) =50%.

a) Find E(T), the average total French mark. [35 points]


b) The lecturer when looking at the distributions of the both the oral and the written
marks noticed that the Var(R) and Var(W) were both higher this year than other years.
Does this automatically imply that the variance of the total mark VAR(T) must also
be higher? Explain why/why not. [65 points]

4. Suppose a researcher is interested in determining the factors that affect crime in the US.
She has observations on crime rates in each state in the US, the % of people living in poverty
in each state and the median age in the state.
In particular, using this state level data and OLS, she estimates the parameters in this simple
regression below.

crime rate= β1 + β 2 poverty rate + β3 medianage +ε

crime rate : crime rate as proportion of population in a state


poverty_rate: Proportion of population living below the poverty line in the state
medianage Median Age in the state. [ Median age is the “middle” age in a state, i.e. 50% of
people in a state are below and 50% of people are above the state’s median age]

The results are shown below and _cons denotes constant. Use this to answer the questions
below.

Page 2 of 3
. regress crime_rate poverty_rate medianage

Source SS df MS Number of obs = 50


F(2, 47) = 16.64
Model .003516345 2 .001758172 Prob > F = 0.0000
Residual .004966951 47 .00010568 R-squared = 0.4145
Adj R-squared = 0.3896
Total .008483296 49 .000173128 Root MSE = .01028

crime_rate Coefficient Std. err. t P>|t| [95% conf. interval]

poverty_rate .1867357 .0375127 4.98 0.000 .1112699 .2622015


medianage -.0016417 .0006132 -2.68 0.010 -.0028754 -.0004081
_cons .0940742 .0242167 3.88 0.000 .0453565 .1427918

a) Comment on the main results (magnitude, sign and significance of coefficients and
Goodness of Fit) in the table. Is this what you would expect, why/why not?
[40 points]
b) Using the results above, predict the crime_rate for a state with a poverty_ rate
=0.15 and a medianage =40. [15 points]

poverty_rate, β 2. You can round your answer to two decimal points.


c) Using a 5% level of significance, test the following for the true coefficient on

H0: β 2=0.20
H1: β 2< 0.20 .
Interpret your results. [25 points]

d) Suppose you wanted the researcher to test whether (in addition to the variables used
above) the crime rate also varied by whether the state governor was Democrat or
Republican. How should the researcher test for this? Be specific in your
instructions. [20 points]

Page 3 of 3

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy