Prof. Dr. Dirk Sliwka and Sander Kraaij
- The sessions take place on Mondays (14:00-17:45) and Tuesdays (09:00-11:45) during the first term of the semester.
- Lecture room: 101 Seminarraum 0.224 (WiSo building)
- Note that we start an hour later on Tuesday than announced on KLIPS.
- We will be using Google Colab throughout the course.
- You can find information about this service here.
- Please bring a laptop (or tablet with a keyboard) to all sessions because we will interactively work with Python in Google Colab.
- Since Google Colab requires an internet connection, please make sure you have a working connection at the university.
- Google Colab also requires a Google account. If you do not have an account, you can create one for free. You do not need to add any personal information to your account to use Colab.
Note: Content will become available sequentially as the time of the session approaches.
- Part 0: Introduction to Python - Introduction - Quiz - Solutions
- Part 1: Regressions - Slides - Notebook LPP - Notebook Management Practices
- Part 2: Statistical Tests - Slides - Notebook Sample Simulation
- Part 3: Causality - Slides - Notebook Sales Simulation
- Part 4: Survey Data, Scale Reliability & Common Method Bias - Slides - Notebook LPP Engagement
- Part 5: Panel Data - Slides - Notebook Panel Data Simulation
- Part 6: Predictions and Machine Learning - Slides Update: Notebook Engagement Predictions - Notebook with Decision Tree Example
- The exam consists of group work and a multiple-choice exam. Please register until October 15!
- Group work:
-
At the end of the second week of the semester, you will be assigned to a group. We will randomly allocate students to groups during the course.
-
In the group, you will analyze a data set of your own choice and apply the methods learned in class. This assignment is designed to help students apply the skills learned in the course to a practical application of their choice.
-
Some hints on how to proceed in your group work
-
Tips for data sets:
-
The groups will give short presentations of their work (15 minutes per group + 10 minutes of discussion) on November 18-19 during the scheduled lecture slots. Groups must hand in their slides and a Juypter Notebook with their analysis code, explanation, and interpretation by November 18 at noon. Each group will be assigned to a time slot to present without the presence of the other groups.
-
- Multiple-choice exam: There will be two exam dates, but we encourage you to take the first one (registration until October 15)
- The 60-minute exam will consist of 15 multiple-choice questions
- Example exam questions
- First Exam: November 30, 2024, 13:00-14:00 at 100 Hörsaal II (main building)
- Second exam: March 12, 2025, 14:00-15:00 at 100 Hörsaal XVIIa (main building)
- Registration for the exam takes place as usual on KLIPS. If you are a doctoral student and do not have access to the course on KLIPS, please write to kraaij[at]wiso.uni-koeln.de that you want to take the exam.
- Examination details
The lecture and tutorial materials we provide will enable you to prepare for the exam. None of the sources below are compulsory, but they might be helpful in addition, in particular:
- Angrist and Pischke Mostly Harmless Econometrics: An Empiricist's Companion (Ch. 2 and 3)
- Arthur Turell's Coding for Economists and Python for Data Science
- James, Witten, Hastie, Tibshirani An Introduction to Statistical Learning with Applications in Python
Further sources:
- Mastering Metrics: The Path from Cause to Effect [Angrist, Pischke]
- Scott Cunningham's Causal Inference - The Mixtape
- Matheus Facure Alves’s Causal Inference for The Brave and True
- Introduction to machine learning with Python: a guide for data scientists [Müller, Guido]
- Introductory Econometrics: A Modern Approach [Wooldridge]
- Andrea Ichino's lecture slides (for some links to standard econometrics courses)