Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

Applied Economics ECONM1008

Part 1 : Microenterprise Training (50 points)

Many people in low-income countries work in the informal sector, in which microenterprises are ubiquitous. Microenterprises usually generate low profits and have only few if any employees. Making microenterprises more profitable could potentially transform them into an engine of growth and employment. One idea is that better managerial know-how and practices might increase their profitability and operational scale. In a study published in the American Economic Journal: Applied Economics, the economists Wyatt Brooks, Kevin Donovan, and Terence R. Johnson test whether this idea could work. The article is available on Blackboard. You do not need to read the full article, but you might find that especially sections one, two, and three (and the introduction) are helpful for answering the tasks.

Task 1.1: What is the research question and hypothesis the authors are testing? (5 points) [at most 100 words]

Task 1.2: Using Stata, replicate Table 3 (OLS Estimates on Profit at Different Time Periods (ANCOVA), p. 207) with the data provided on Blackboard. Section III.A provides more details on the specification. Use a global to specify the set of control  variables. Use esttab and export the regression table to Word. Your table should look like the original one, but you can simply use numbers as column titles. Please copy and paste the relevant Stata code and the table it produces below. Do    your results differ from the results published in the original paper? If so, how? (15 points)

Task 1.3: Interpret column 1 of Table 3 (use the estimates published in the paper), with particular emphasis on (i) the marginal effects, (ii) whether the effects can be interpreted in a causal way, and (iii) whether the coefficients allow us to identify the effect of actually having attended the business class and having interacted with a mentor (or just an intend- to-treat effect). (15 points) [at most 250 words]

Task 1.4: Using Stata, replicate Figure 3 (Profit Time Series, p. 206) with the data provided on Blackboard. Your figure does not need to include the grey rectangle highlighting when the intervention took place. Your legend can also be below the figure. Please copy and paste the relevant Stata code and the figure it produces below. How does your figure differ from the figure published in the original paper? (10 points)

Task 1.5: A policymaker in Bangladesh sees the results and wants to introduce a mentorship scheme for male microentrepreneurs in Bangladesh. What would be your advice? (5 points) [at most 150 words]

Part 2 : Cycling to school (36 points)

Ensuring inclusive and equal access to education is high on the global policy agenda. While girls have caught up and even overtaken boys in many countries, especially at the primary school level, girls still lack behind in many low- and middle-income countries, especially in secondary schooling and higher education. Policymakers have tried different tools to increase girls’ enrolment in schooling. Some of these policies aim at increasing demand for schooling by providing cash transfers to families conditional on their offspring attending schools, and other policies have tried to increase the supply of schools by constructing more schools. Building new schools is costly, and an alternative way to improve access is to help students with transportation. In 2006 the Indian state of Bihar therefore gave all girls who enrolled in grade 9 means to buy a bicycle, which could help them access the school. In a study published in the American Economic Journal: Applied Economics, the economists Karthik Muralidharan and Nishith Prakash investigate whether this policy worked. The article   is available on Blackboard. You do not need to read the full article, but you might find that especially sections one and two are helpful for answering the tasks.

Task 2.1: Formulate the evaluation problem using the potential outcome framework: (i) Define the unit of observation, (ii) define the treatment D, (iii) define the outcome variable(s) of interest Y, (iv) explain which two potential outcomes this variable can take on for each unit i, (v) explain which of the two potential outcomes is observed for each unit. (12 points)  [at most 150 words]

Task 2.2: The study uses the difference-in-differences approach to identify the causal effect of the cycle program. Using your own words, explain why we cannot use the increase in female secondary school enrolment from before to after 2006 in Bihar to conclude that the cycling program had a positive effect. Explain how the difference-in-differences approach is able to address some of the problems of the simple before-after comparison. (12 points) [at most 150 words]

Task 2.3: Muralidharan and Prakash use two control groups: boys and girls in other states of India. They do so, because they believe that only using boys as a control group leads to a violation of an important assumption behind the difference-in-differences approach. Explain what this assumption is and how the results in Panel A of Table 1 (Testing the Parallel Trends Assumption, p. 330) suggest that this assumption would be violated if only using boys as the control group. (12 points) [at most 150 words]

Part 3 : Summer School and Test Scores (R script) (14 points)

The following script follows the example used in the R tutorial that we created for you (see https://hhsievertsen.github.io/applied_econ_with_r/), where we load and analyze fictitious data on test scores and summer school attendance and child background.

Task 3.1: In the code block below we run a regression of summer school attendance on controls for student background  and an indicator for receiving the reminder letter. Replace XYZ1, XYZ2, and XYZ3 in the code block below with the correct code and interpret the results. Your response should start with a list of what you replaced the three elements with (i.e.,   XYZ1=…, XYZ2=…, XYZ3=…). Your response should include the R output. (5 points) [at most 100 words, not counting the R output and the list of replaced terms]

# Estimate LPM (the first stage)

models<-list (

m1<- XYZ1 (summerschool XYZ2 letter,cluster="school_id",data=regdata),

m2<- XYZ1 (summerschool XYZ2 letter+parental_schooling+parental_lincome XYZ3 female,cluster="school_i

d",data=regdata)

)

# Store the mean of dependent variable in a data frame

added_stats<-tibble ("Mean of Dep . ",m1=mean (regdata$summerschool),m2=mean (regdata$summerschool))

# Generate table

modelsummary (models, stars = TRUE,statistic = 'std .error',

fmt= '% .4f',add_rows = added_stats,

coef_omit= '(Intercept)', output = 'flextable')

Task 3.2: A policymaker hears about your results in Task 3.1 and states that “The analysis by a Bristol economist shows a clear positive significant treatment of the summer school.” Explain why this statement is not correct. Because you are a helpful economist you provide some additional analysis to inform the policymaker. Replace XYZ1, XYZ2, and XYZ3 in the code block below with the correct code and interpret the results and explain how these results provide insights into the claim made by the policymaker. Your response should include the R output. (5 points) [at most 100 words, not counting the R output and the list of replaced terms]

# Ordinary Least Squares regression

model1<-lm(test_score~parental_schooling+parental_lincome+letter+female, XYZ1= XYZ2 (analysisdata,year=

=6))

# Summary of model1

XYZ3 (model1)

Task 3.3: Replace XYZ1 and XYZ2 in the R code below and interpret the findings. Your answer should include the R output.

(4 points) [at most 100 words, not counting the R output and the list of replaced terms]

# Estimate IV specification with feols

m1<-feols (test_score~parental_lincome+female+parental_schooling | # Outcome eq .

0 | # Fixed effects

XYZ1~ XYZ2 # First stage

,cluster="school_id" # Cluster var

,data=regdata)

# Summary of results

summary (m1)

Penalties for late work

Assignments handed in after the deadline, without a pre-arranged extension will be subject to the following penalty:

A fixed absolute penalty of 10 marks is applied for each day work is submitted after the agreed submission deadline. Please note, weekend days count towards the calculation of late penalties, bank holidays and University closure days do not.

A mark of zero is applied to work submitted five or more days after the agreed deadline if this threshold is not already reached.

Plagiarism

In academic writing, plagiarism is the inclusion of any idea or any language from someone else without giving due credit by citing and referencing that source in your work. This applies if the source is print or electronic, published or unpublished, another student’s work, or any other person.

The University's Examination Regulations state that Any thesis, dissertation, essay, or other course work must be the students own work and must not contain plagiarised material. Any instance of plagiarism in such coursework will be treated as an offence under these regulations.” (Section 3.1).

The Examination Regulations give information on the University's procedures for