Computational Economics I und II

Computational Economics I (Weeks 1-8)

Computational methods are of crucial importance for applications in economics and beyond. This ranges from economic modelling to applied economics to econometrics to finance and data science. This seminar intends to give a broad overview over the techniques, methods, and applications. It includes programming with the general purpose programming language (Python) and demonstrates how computational methods are implemented. It allows students to gain experience with the simple applications of computational methods in general and with the programming language primarily used in the course (Python) in particular. While the economic application of sophisticated methods like machine learning, agent-based modelling, and micro-econometrics, and natural language processing is beyond the scope of this course, it will give students a basic understanding and a good starting point for further studies in these fields.

Computational Economics II (Weeks 9-17)

Building on the course Computational Economics I, this course covers a selection advanced methods of computational economics and their application to cases in economics, but also finance, business or other social sciences. This includes in particular, agent-based modeling, natural-language processing, data cleaning, general statistics, as well as some additional techniques of machine learning and visualization. Conceptual and practical problems such as limits to computation time, the curse of dimensionality, non-reproduibility, incomplete data and data availability issues will also be discussed and applied in the context of examples from economics. As such, the course provides a deep dive into a field that is rapidly becoming one of the crucial methods on which research in applied economics builds.The course uses the programming language Python. Students will be required to work with programming and perform data analysis on new data sets, applying methods discussed in the course.

There will be no examination. Grades will be based on homework assignments which have to be completed during the term.

Computational Economics I

1. Introductory lecture: Computational economics and Python
2. Basic programming techniques and data structures
3. Programming style and good practice
4. Computational techniques: Visualizing data
5. Computational techniques: Statistical and econometric analysis
6. Computational techniques:Simulations
7. Computational techniques: Network theory
8. Outlook: Text mining, machine learning, agent-based modelling

Computational Economics II

1. Introductory lecture and recap of the contents of Computational Economics I
2. Model analysis techniques: Experiments and testing
3. Model analysis techniques: Simulation
4. Model analysis techniques: Verification
5. Model analysis techniques: Validation and calibration
6. Computational science techniques: Agent-based modeling
7. Computational science techniques: Small and large data sets; data cleaning and processing
8. Computational science techniques: Classification methods in machine learning
9. Computational science techniques: Natural-language processing
10. Outlook

Recommended literature:

Wentworth, P., Elkner, J., Downey, A. B., and Meyer, C. (2012). How to think like a computer scientist: Learning with Python 3. Open Book Project: http://openbookproject.net/thinkcs/python/english3e/index.html

Complementary literature:

Elsner, W., Heinrich, T., and Schwardt, H. (2015). Microeconomics of Complex Economies: Evolutionary, Institutional, Neoclassical, and Complexity Perspectives. Academic Press, Amsterdam, NL, San Diego, CA, et al.

Hommes, C., LeBaron, B. (2018). Computational Economics: Heterogeneous Agent Modeling. Volume 4, The Netherlands North-Holland, Amsterdam.

Sargent, T. and Stachurski, J. (2020). Quantitative economics with Python. Online version: https://python.quantecon.org/