MP223 Applied Econometric Methods for the Social Sciences

Justus Liebig University Giessen · Summer Semester 2022

Lecture slides, R code and project bundles

Everything developed for MP223 in one place: seven topics, fifteen lecture decks, twenty worked R exercises and seven downloadable RStudio projects. Start with the slides below, then work through the matching exercises.

Taught and written by Eduard (Eddie) Bukin · Ilias · StudIP · YouTube playlist

Lecture slides

All fifteen decks, in teaching order. Every deck opens as a full-screen presentation — press ? inside it for navigation shortcuts.

1 · Introduction

Introduction. Organisation. Setup.

How the course runs, software setup, and the two ideas everything rests on: ceteris paribus and selection bias.

2 · Data wrangling

Workflow, tidy data, wrangling

The data-analysis workflow and what makes a dataset tidy.

2 · Data wrangling

Wrangling with dplyr

Extended version of the workflow deck, with the full dplyr verb set worked through.

2 · Data wrangling

Import data

Reading CSV and Excel files with readr and readxl, and cleaning names with janitor.

3 · Describing data

Exploring numerical data

Distributions, summaries and the plots that reveal them.

3 · Describing data

Correlation

Correlation types, how to read their strength, and where they mislead.

4 · Simple regression

Simple regression

One predictor, least squares, and what the coefficients actually mean.

5 · Multiple regression

Multiple linear regression

Several predictors at once, and holding other things equal in practice.

5 · Multiple regression

Linearity

Diagnosing non-linearity and fixing it with transformations.

5 · Multiple regression

Omitted variable bias

What leaving a variable out does to the estimates you keep.

5 · Multiple regression

Collinearity

When predictors carry the same information, and how to detect it.

5 · Multiple regression

Heteroscedasticity

Non-constant error variance, its consequences, and robust standard errors.

5 · Multiple regression

Hedonic prices model

A full applied example: decomposing prices into the value of characteristics.

6 · Panel regression

Panel regression analysis

Fixed and random effects, and what repeated observations buy you.

7 · Instrumental variables

Instrumental variable

Identification when a regressor is endogenous, and the returns-to-schooling example.

Work through the code

Every exercise is published as a rendered walkthrough you can read in the browser, and shipped inside an RStudio project you can download and run.

Exercises

All exercises & code

Twenty worked R walkthroughs, grouped by topic, with solutions where they exist.

Bundles

Downloadable projects

Seven ready-to-open RStudio projects with their data. See what is inside before downloading.

Resources

External learning resources

Books, cheat sheets, interactive tutorials and videos for learning R.

How the materials are labelled

The same icons are used in the sidebar and on every topic page:

Icon	Material
🖥	Lecture slides — a revealjs deck, opens full screen
📋	Application exercise — a rendered R walkthrough
📦	Project bundle — a downloadable RStudio project with data
📺	Video recording
📖	Reading or cheat sheet

About this site

This site collects the lecture slides, R application exercises and supplementary materials developed for MP223 Applied Econometric Methods for the Social Sciences at JLU Giessen, Summer Semester 2022. It does not replace the Ilias or StudIP course pages — it makes the slides and code accessible, downloadable and readable on any device.

Written and taught by Eduard (Eddie) Bukin — applied economist and open-source data scientist. All materials are open source and available on GitHub.