leandrozipitria / empirical_io_course

Material from my master level course "Empirical Industrial Organisation and Consumer Choice"

Empirical Industrial Organisation and Consumer Choice

Sebastian Kranz, Ulm University

This Github repository contains material from the 2018 edition of my master level course "Empirical Industrial Organisation and Consumer Choice" that I teach regularly at Ulm University.

We have quite a heterogeneous composition in the class and not every student has heard yet an econometrics class. Hence, the course starts at a very basic level and ends already at the famous BLP model that allows to estimate demand functions for differentiated products.

We also have a mini-excursion to machine learning in exercise sheet 4 and 5. Main goal is to contrast the approach to pure prediction problems of machine learning with the estimation of causal relationships with structural econometric models.

If you like R programming, take a look at the folder rtutor that contains RTutor problemsets, that allow you to explore the contents of the early chapters in an interactive fashion. The RTutor problemsets 1a to 1c mainly deal with the question of how we can estimate a very simple demand function from field data. You learn very slowly with simple Monte-Carlo simulations about endogeniety problems in this context and how one can attempt to solve them.

See https://github.com/skranz/RTutor for more details about RTutor and links to other problem sets.

To install the required R packages for this course, run the code in install packages.r in the r folder.

The folder slides contains some lecture slides. Often I do some live programming in the class. The code is either part of the RTutor problem sets (early slides) or in the folder r (later slides). Alternatively, I show details on the black board. So stand-alone the slides have only limited use.

The folder `exercises`` contains the problem sets and sample solutions for the exercise classes of the course. These are additional problem sets to the RTutor problem sets.

License

• Except for the RTutor problem sets, you can use all materials under a Creative Commmons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) license license.

• You can also use and share the RTutor problem sets similar to the (CC BY-NC-SA 4.0) license with one restriction: You are not allowed to use or share the problem sets if you make or plan to make publicly available or provide a link to any solutions of these problem sets. (To motivate students to solve the problem sets, they are grade-relevant to a small extend. Therefore I don't want solutions too easily available.)

Sources & Attributions

• The exercises and sample solutions in the exercise folder have been created by Martin Kies with some input by me.

• The data folder contains a set about car sales, which was made available on Frank Verbooven's website.

• The slides EmpIO_2b.pdf and the corresponding R code is based on this Vignette of the R package mlogit by Yves Croissant who has adapted the R code from original Matlab exercises by Kenneth Train. (The R package is distributed under a GPL License)

• Exercise Sheet 3 is an adaption from a problem set by Glenn Ellisions Industrial Organisation Course at the MIT, which is available as part of MIT-OCW under a Creative Commmons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) license.