Introduction to Econometrics
& Programming

Material

Fall semester

		Slides	Quizzes	Data
Part 1: Introduction to R programming
Lecture 1	Introduction to R	html - pdf	Quiz 1	data
Lecture 2	Descriptive statistics	html - pdf	Quiz 2
Lecture 3	Basic data manipulation	html - pdf	Quiz 3	data
Lecture 4	Data visualization	html - pdf		data
Lecture 5	Quarto & \(\LaTeX\)	html - pdf		data
Lecture 6	Text data & sentiment analysis	html - pdf		data
Lecture 7	Homework correction	html - pdf		data
Part 2: Introduction to Econometrics
Lecture 8	Univariate regressions	html - pdf	Quiz 4	data
Lecture 9	Multivariate regressions	html - pdf	Quiz 5	data
Lecture 10	Inference	html - pdf	Quiz 6	data
Lecture 11	Causality	html - pdf
Lecture 12	Interpretation	html - pdf
Lecture 13	Applications in academic research	html - pdf		data
Lecture 14	Maps and geolocalized data	html - pdf		data

	Guidelines	Source	Data
Exams
Homework	html - pdf	qmd	data

Spring semester

		Slides	Data
Guidelines and refreshers
Lecture 16	How to conduct a research project	html - pdf
Lecture 17	Refresher: R Programming	html - pdf	data
Lecture 18	Refresher: Econometrics	html - pdf

	Guidelines	Grading	Data
Exams
Research project	html - pdf	html - pdf	data

Cheatsheets

Contents

Fall semester

I. Introduction to R

1. Getting started

• 1.1. About R

• 1.2. The R Studio IDE

• 1.3. Import and eyeball data

• 1.4. Use functions

2. Anatomy of a data.frame

• 2.1. Data structure

• 2.2. Classes

• 2.3. Vectors

• 2.4. Subsetting

• 3. Wrap up!

II. Descriptive statistics

1. Distributions

• 1.1. Definition

• 1.2. Graphical representation

• 1.3. Common distributions

2. Central tendency

• 2.1. Mean

• 2.2. Median

• 2.3. Mean vs. median

3. Spread

• 3.1. Range, quantiles, and the IQR

• 3.2. Variance and standard deviation

• 3.3. Standard deviation vs. interquartile range

4. Inference

• 4.1. Data generating process

• 4.2. Empirical vs. theoretical moments

• 4.3. Confidence interval

• 5. Wrap up!

III. Basic data manipulation

1. The dplyr package

• 1.1. Packages

• 1.2. Basic functions

• 1.3. group_by() and summarise()

2. Merge and reshape

• 2.1. Merge and append data

• 2.2. Reshape data

3. A few words on learning R

• 3.1. When it doesn't work the way you want

• 3.2. Where to find help

• 3.3. When it doesn't work at all

• 4. Wrap up!

IV. Data visualization

1. The ggplot() function

• 1.1. Basic structure

• 1.2. Axes

• 1.3. Theme()

• 1.4. Annotation

2. Adding dimensions

• 2.1. More axes

• 2.2. More facets

• 2.3. More labels

3. Types of geometry

• 3.1. Points and lines

• 3.2. Barplots and histograms

• 3.3. Densities and boxplots

4. How (not) to lie with graphics

• 4.1. Cumulative representations

• 4.2. Axis manipulations

• 4.3. Interpolation

• 5. Wrap up!

V. Quarto & LaTeX

1. Basic principles

• 1.1. What is Quarto?

• 1.2. YAML header

• 1.3. Code chunks

• 1.4. Text formatting

• 1.5. Run and render your code

2. Useful features

• 2.1. Inline code

• 2.2. Tables

• 2.3. Preset themes

• 2.4. Report parameters

3. LaTeX for equations

• 3.1. What is LaTeX?

• 3.2. LaTeX syntax

• 3.3. Large equations

• 4. Wrap up!

VI. Text data & sentiment analysis

1. Cleaning text data

• 1.1. Exploring the data

• 1.2. Regular expressions

• 1.3. Tokenization

2. Sentiment analysis

• 2.1. Stopwords

• 2.2. Sentiments

• 2.3. Analysis

• 3. Wrap up!

• VII. Homework correction

VIII. Univariate regressions

1. Joint distributions

• 1.1. Definition

• 1.2. Covariance

• 1.3. Correlation

2. Univariate regressions

• 2.1. Introduction to regressions

• 2.2. Coefficients estimation

3. Binary variables

• 3.1. Binary dependent variables

• 3.2. Binary independent variables

• 4. Wrap up!

IX. Multivariate regressions

1. Adding variables

• 1.1. Continuous variables

• 1.2. Discrete variables

2. Control variables

• 2.1. Motivation

• 2.2. Discrete

• 2.3. Continuous

3. Interactions

• 3.1. Motivation

• 3.2. Discrete

• 3.3. Continuous

• 4. Wrap up!

X. Inference

1. Asymptotic inference

• 1.1. Data generating process

• 1.2. Standardization

• 1.3. Confidence interval

2. Exact inference

• 2.1. Standard error

• 2.2. Student-t distribution

• 2.3. Confidence interval

3. Hypothesis testing

• 3.1. P-value

• 3.2. linearHypothesis()

• 4. Wrap up!

XI. Causality

1. Main sources of bias

• 1.1. Omitted variable bias

• 1.2. Functional form

• 1.3. Selection bias

• 1.4. Measurement error

• 1.5. Simultaneity

2. Randomized control trials

• 2.1. Introduction to RCTs

• 2.2. Types of randomization

• 2.3. Multiple testing

• 3. Wrap up!

XII. Interpretation

1. Point estimates

• 1.1. Continuous variables

• 1.2. Discrete variables

• 1.3. Log vs. level

• 2. Practice interpretation

3. Regression tables

• 3.1. Layout

• 3.2. Reported significance

• 3.3. R squared

• 4. Wrap up!

XIII. Applications in academic research

1. Causal approach (Behaghel et al., 2015)

• 1.1. Structure

• 1.2. Data

• 1.3. Analysis

2. Correlational approach (Chetty et al., 2014)

• 2.1. Empirical approach

• 2.2. National results

• 2.3. Spatial variations

• 2.4. Correlational analysis

3. Structural approach (Nerlove, 1963)

• 3.1. Motivation

• 3.2. Theoretical modeling

• 3.3. Regression expression

• 4. Wrap up!

XIV. Maps and geolocalized data

1. Geolocalized data

• 1.1. Shapefiles and rasters

• 1.2. Opening geolocalized data

• 1.3. Coordinate Reference Systems

• 1.4. Subsetting geolocalized data

2. Geographic variables

• 2.1. Import from csv

• 2.2. Zonal statistics

• 2.3. Centroids and distances

• 3. Wrap up!

Spring semester

I. How to conduct a research project

1. Preliminary steps

• 1.1. Research question

• 1.2. Finding data

• 1.3. Literature review

2. Data description

• 2.1. Data cleaning

• 2.2. Descriptive statistics

• 2.3. Data visualization

3. Analysis

• 3.1. Regression analysis

• 3.2. Robustness

• 3.3. Heterogeneity

II. R Programming & Descriptive statistics

1. The basics of R programming

• 1.1. Types of R objects

• 1.2. The dplyr grammar

• 1.3. Data visualization

2. Descriptive statistics

• 2.1. Distributions

• 2.2. Central tendency

• 2.3. Spread

• 2.4. Joint distributions

3. A few words on using R

• 3.1. When it doesn't work the way you want

• 3.2. Where to find help

• 3.3. When it doesn't work at all

III. Introductory Econometrics

1. Regressions with continuous variables

• 1.1. Estimation

• 1.2. Inference

2. Regressions with discrete variables

• 2.1. Binary dependent variable

• 2.2. Binary independent variable

• 2.3. Categorical independent variable

• 3. Controls and interactions

• 4. Interpretation

Syllabus

Course description

The objective of this course is to provide you with the necessary statistical and data visualization tools to perform well-structured and meaningful data analyses for your own research projects. By the end of this course you should be able to produce relevant statistics and compelling graphics to include in your future reports, presentations, or research projects. To this end, this course combines notions of R programming, basic Statistics, and introductory Econometrics.

During the first semester, the emphasis will successively be put on R Programming and Econometrics. Still, all along the course bridges will be built between these two dimensions of data analysis. The second semester will be dedicated to a research project. You will have to carry out an original data analysis by applying the tools covered in class during the first semester to your own research question. For your soon-acquired programming and data analysis skills to be easily valuated on an online CV or as a writing sample, you will learn how to make your final project look like this document.

Practical matters

The course will be taught in French but the material is written in English. It is meant to be accessible without prior knowledge in R programming or Statistics. Lectures will take place at Campus Jourdan - Room 3-38, every Monday from 13:30 to 15:30. Please reach me at louis.sirugue@psemail for any question or comment about the course.

Please make sure to download and install R, RStudio, and Quarto before the first lecture, and to bring your laptop with you at each class.

• Download R
• Download RStudio
• Download Quarto

Don’t hesitate to send me an email if you’re already facing issues at that stage.

Grading system

Fall semester

The grading of the first semester will be based on:

• Six weekly online quizzes - 25%
• One homework - 30%
• One final exam - 45%

Online quizzes consist in sets of 3 to 5 short questions related to the content of the previous class. Links to the quizzes will be available in the Material section of the course webpage. You will be able to log in to each quiz using the verification code received by email. You will have the possibility to retry the quiz as many times as you want before submitting. When you’re done with a quiz you must click on download results and send the downloaded file to me by email. This is to keep track of your answers in case of disagreement about the grading. You have until the beginning of the following lecture to submit your results and send them to me by email.

The homework is a set of exercises related to the first five lectures of the course. It can be done alone or by pairs, and must be handed over via email by the end of the sixth week (15^th of October 18:00). Late submissions will be penalized by half a point for each 30min beyond the deadline. It will soon be available on the Material section of the course webpage. The precise grading system of the homework is also available on the Material section of the course webpage already. You are very welcome to help each other, but you must write your answers yourselves. Copy-pasting is not permitted.

The final exam will be a paper exam in classroom. You are allowed to bring a cheatsheet with you to the exam as long as it is handwritten and standing on a single A4 (21cm x 29.7cm) page, i.e., recto only. Cheatsheets that do not comply with these rules will be confiscated. Beyond that, standard examination rules apply.

Spring semester

The grading of the second semester will be entirely based on your research project that you will carry out all along the semester. Research projects must be done by groups of two, so both members of a pair will have the same grade. You will be evaluated three times, at three different stages of the research process:

• Presentation of research idea - 25%
• Midterm report - 30%
• Final project/presentation - 45%

As the first three lectures of the second semester will be refreshers on the notions seen during the first semester, you will have some time to come up with your own research question and to find relevant data to test your hypotheses empirically (you will be guided on where and how to look for data). Then, in each session you will have to apply a given step of the research process to your research project: data cleaning, descriptive statistics, data visualization, hypothesis testing, etc. until your final presentation. Here is an example of what the final document should look like, and the guidelines and grading system are available in the Material section of the homepage. You will have the possibility to write it in French or in English depending on what you think would be the most useful for you.

Schedule

Fall semester

Part 1: Introduction to R programming
04/09/23	Lecture 1 *	Introduction to R
11/09/23	Lecture 2 *	Descriptive statistics
18/09/23	Lecture 3 *	Basic data manipulation
25/09/23	Lecture 4	Data visualization
02/10/23	Lecture 5	R markdown & \(\LaTeX\)
09/10/23	Lecture 6	Text data & sentiment analysis
16/10/23	Lecture 7	Homework correction
Part 2: Introduction to Econometrics
23/10/23	Lecture 8 *	Univariate regressions
06/11/23	Lecture 9 *	Multivariate regressions
13/11/23	Lecture 10 *	Inference
20/11/23	Lecture 11	Causality
04/12/23	Lecture 12	Interpretation
11/12/23	Lecture 13	Applications in academic research
18/12/23	Lecture 14	Maps and geolocalized data
15/01/24	Lecture 15	Final exam

Standard lectures:

During these lectures we will cover the material core material of the course.

Buffer lectures:

These lectures are slightly more advanced, but their content will not be needed for the exams. There are two buffer lectures, each placed before an exam to give you time to prepare for the exam.

Exams:

The homework is due by the 15^th of October 18:00, and will be corrected in class during the 7^th lecture.
The final exam will take place during the very last lecture.
The six online quizzes will take place during the first three weeks of each part of the course. The corresponding lectures are marked with a * symbol.

Horizontal dashed lines represent break periods. They all last 1 week except the 3-week-long Christmas break.

Spring semester

	Group 1	Group 2
Part 1: Guidelines and refreshers
Lecture 1	TBA	TBA	How to conduct a research project
Lecture 2	TBA	TBA	Refresher: R Programming
Lecture 3	TBA	TBA	Refresher: Econometrics
Part 2: Research project
Lecture 4	TBA	TBA	Presentation of your research question and dataset
Lecture 5	TBA	TBA	Follow-up: Data cleaning I
Lecture 6	TBA	TBA	Follow-up: Data cleaning II
Lecture 7	TBA	TBA	Follow-up: Descriptive statistics
Lecture 8	TBA	TBA	Follow-up: Visualizing the data
Lecture 9	TBA	TBA	Follow-up: Regression analysis
Lecture 10	TBA	TBA	Follow-up: Midterm report feedback
Lecture 11	TBA	TBA	Follow-up: Causality assessment
Lecture 12	TBA	TBA	Follow-up: Robustness
Lecture 13	TBA	TBA	Follow-up: Heterogeneity
Lecture 14	TBA	TBA	Follow-up: Last tips
Lecture 15	TBA	TBA	Final presentation

Guidelines and refreshers:

The first lecture will cover in detail how to conduct your research project, and the next two lectures will go back on the most important notions seen during the first semester. These are meant to give you time to form groups of 2 and to find your research question and your dataset. You will be guided on where and how to look for data, and both the data and the research question can be taken from an existing academic article.

Follow-ups:

Follow-up sessions are meetings during which you will be given feedback on what you’ve done and indications on what to do next. Between each session you will have to apply a given step of the research process to your research project: data cleaning, descriptive statistics, data visualization, hypothesis testing, etc., until your final presentation. Here is an example of what the final document should look like, and the guidelines and grading system are available at the bottom of the Material section of the homepage. You will have the possibility to write it in French or in English depending on what you think would be the most useful for you.

Exams:

The 4^th lecture will be dedicated to the presentation of your research questions and datasets.
A midterm report must be handed out by email after the 9^th lecture. Lecture 10 will be devoted to giving you feedback about it.
The last lecture will be dedicated to your final presentations.
More detailed guidelines are provided above and in the first set of slides “How to conduct a research project”.

Horizontal dashed lines represent break periods.

References

This course is inspired by:

• Introduction to R, by H. Bull and P. Charousset. Paris School of Economics (2020)
• Advanced Mircoeconometrics, by D. Margolis and F. Libois. Paris School of Economics (2020)
• How to Lie with Graphics, by C. Bontemps. Toulouse School of Economics (2020)
• Introduction to Econometrics with R, by F. Oswald, V. Viers, P. Villedieu, and G. Kenedi. SciencesPo Dep. of Economics (2020)
• Introduction to Causal Inference, by L. Zabrocki. Paris Sciences et Lettres - CPES (2020)
• Geolocalized Datasets and Applications for Economics, by F. Libois and E. Madinier. Paris School of Economics (2020)
• Countless posts on stackoverflow and R-bloggers