Scribd
Upload
26 views2 pages

Stats Mid-Term Exam

The document outlines a mid-term examination for an Advanced Statistics course, consisting of theoretical questions and applied problems related to hypothesis testing, regression analysis, and program evaluation. It includes specific instructions for hypothesis formulation, statistical calculations, and interpretation of results. The exam covers topics such as OLS estimators, public health program evaluation, and wage analysis, with a total duration of 2 hours and a maximum score of 20 points.

Uploaded by

oussama fq
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
26 views2 pages

Stats Mid-Term Exam

The document outlines a mid-term examination for an Advanced Statistics course, consisting of theoretical questions and applied problems related to hypothesis testing, regression analysis, and program evaluation. It includes specific instructions for hypothesis formulation, statistical calculations, and interpretation of results. The exam covers topics such as OLS estimators, public health program evaluation, and wage analysis, with a total duration of 2 hours and a maximum score of 20 points.

Uploaded by

oussama fq
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/ 2

Mid-ternm Examination

Advanccd Statistics - L3

Duration: 2 hours; total points: 20

Note:

" For hypothesis testing, you should (i) write the hypotheses using proper mathematical notation, (ii) show all
steps in your t-statistic calculation, (iii) clearly state your conclusion
" Critical valuesare 1.645 for 10% significance level, 1.96 for 5% significance level and 2.576 for 1% significance
level

Part I: Theoretical Questions (6 points)


1. (2 points) List and explain:
(a) The assumptions needed to obtain an unbiased OLS estimator
(b) The additional assumptions needed for OLS to be BLUE
2. (2 points) Consider the estimated variance of B1 in simple linear regression:

-(a) Write down the formula for Var(B1) =


(b) Explain how this variance changes when:
" The sample size increases
. Var(X) increases
" g increases

3. (2 points) In the context of matching:


- (a) What does the conditional independence assumption correspond to?
(b) What is the curse of dimensionality? How do researchers overcome this issue?

Part II: Applied Problems (14 points)


Problem 1: Public Health Program Evaluation (7 points)
Acity is considering providing free gym memberships to residents and wishes to know the potential effects of this
program. You have data on 5,000 residents including their weekly exercise hours and whether or not they have a
gym membership.

1. (1 point) One of the city politician suggests you should study whether, among the 5,000 residents, those
who have a gym membership are exercising more. Write down (i) a regression specification to estimate the
politician's suggestion, using appropriate variables, o. 31. and the error term u, (ii) the hypothesis the
politician wishes to test.
2. (3 points) The city administration implements the politician's suggestion and releases the following results:
erercise_hours = 2.34 + 1.2 gym_membership
(0.32) (0.6) 2,34
n=5,000, R = 0.23 O,32
(Standard errors in parent heses)
1
(n) Test 1he mall lypothesis that gym nenership has no effect on exercise hours against the alternative
hypothesis that it lhas a statistically significant cfect at the 10%, 5% and 1% significance levels.
-(b) Provide the fornmla to find the 95% confidence interval for the gym membership cocfficient.
(e) Do youthink that 1.2 correspond to the causal effect of gym nembership on hours of exercise? Justify
your answer using two different examples of your choice.

8. (3 points) The city decides to evaluate the program using a randomized controlled trial;
(a) What is a randomizcd controlled trial? What are balance checks useda for? to chedk whelhen a test
(ealy tamdsm
-(b) The validity of a randomized controlled trial is often cvaluated using the concepts of internal and external
validity. Recall the definition of these two concepts.
- (c) List two clements that could threaten the internal validity of this study.
Problem 2: Education and WVage Analysis (7 points)
We cstimate two models in order to explain the annual wage. We have the following variables: wage designates
the annual wage, education the number of years of education, ezperience the number of years of work experience,
parentincome the income of the parents of the individual, female a binary variable equal to 1 if the individual is
a woman.
Model 1:
log(wage) = 9.45 +0.16 education
(0.15) (0.02)
n=1,000, R = 0.21
Model 2:
log(wage) = 9.12 +0.08 education + 0.03 ezperience + 0.05 log(parentincome) - 0.2 female
(0.14) (0.02) (0.01) (0.01) (0.04)
n= 1,000, R²= 0.34

(Standard errors in parentheses)

1. (2.5 points) Interpretation:


(a) Interpret the education coefficient in Model 1 and explain why this coefficient changes in Model 2
(b) Interpret the log(parentincome) coefficient in Model 2.
(c) A man shows up with 10 years of education, 5 years of experience, a parental income of 10,000 euroS.
Write down how you would compute the expected annual wage of this individual based on Model 2 (you
do not need to perform the final calculations).
2. (2.5 points) Statistical analysis:
(a) Test whether being a woman has an effect on wages, all else being equal, at the 10%, 5% and 1%
significance levels.
(b) Provide theformula to find the 95% confidence interval for the experience coeficient.
(c) Interpret the R in model 2 and explain the diference in R between the two models.
3. (2 points) Econometric discussion:
(a) Considering model 2: list two factors that might be included in u that could be correlated with education
and influence annual wage
(b) If you had conmplete freedom, explain (i) how you would evaluate the causal effect of education on wages
and (ii) make sure that you are really estimating a causal effect?

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