/empecon

Empirical Economics with R

Primary LanguageHTMLOtherNOASSERTION

Empirical Economics with R

by Sebastian Kranz at Ulm University.

This advanced Bachelor course was first taught in the winter term 20/21 and due to the Corona crisis can be followed as an online course. This Github repository contains all materials or links to it.

Videos, multiple choice quizzes and RTutor problem sets

Besides lecture slides, the course consists of interactive web sites that mix video lectures with multiple choice quizzes. Even more important are interactive RTutor problem sets for each chapter. They allow you to work through the topics and applications in your own RStudio environment. You can automatically check your solutions and get hints. Look at setup.md to get started and install all required packages. The material for each chapter is linked and described below.

Course content

The course consists of the following 5 chapters.

Chapter 1 motivates linear regression with a classic application: developing a formula to predict future prices of Bordeaux Red Wines.

Chapter 2 is a short introduction into Machine Learning including regression trees, random forests and cross-validation. The main example is a classic house price prediction.

Chapter 3 digs deeper into linear regression, focusing on the estimation of causal effects and endogeneity problems. In particular, I try to distill the importance of a source of exogenous variation if we want to have any chance of estimating causal effects.

Chapter 4 covers difference-in-difference estimation. The main example is an experiment by eBay that estimates the effect of search engine marketing on revenues. The RTutor problem set also looks at a classic study on the employment effect of minimum wages.

Chapter 5 covers instrumental variable estimation and the potential outcomes framework for causal identification. The main application is a large randomized experiment that compares the efficiency of public and private job market counselling.