Computational Methods for Economists (CMFE) – summer course

Welcome to CMFE 2019!

There has been an explosion of interest in text analysis within the fields of economics, management, and finance. The “Computational Methods for Economists: Text analysis and agent-based models” course will teach doctoral and master students a range of computational methods in topic modeling, statistical methods for making causal inferences from text and agent-based computational economics. Through a series of hands-on tutorials, the students will also work on real examples with real coding solutions that they can refer back to in future research projects.


EPFL campus with the Alps in the background. Copyright: EPFL – Alain Herzog.

Overview

Computational Methods for Economists: Text analysis and agent-based models

2019 EPFL/ETH summer school

Venue: EPFL Odyssea
Dates: 10-15 June 2019
Hours: 6 days of 3 hours each, with pre-course readings and assignments
Credits: 2

Total workload should be approximately 56 hours: 20h of instruction, 12 hours of pre-reading and 24 hours for exercises (4 hours per exercise). Each course day is split into 2 hours of lectures and 1 hour of Python/R tutorial. Six sessions will have an exercise requiring approximately 4 hours to complete at home. Exercises are to be submitted to the respective professors, 2 weeks after the course end. Grades are pass/fail: R for pass, E for fail.

Click here for the official flyer.

Instructors

Stephen Hansen – Associate Professor in the Economics Department at the University of Oxford (https://sekhansen.github.io/)

Margaret E. Roberts – Associate Professor in the Department of Political Science at the University of California, San Diego (http://www.margaretroberts.net/)

Yaroslav Rosokha – Assistant Professor of Economics at Krannert School of Management at Purdue University (https://web.ics.purdue.edu/~yrosokha/)

Guest lecture

Harsh Prasad – Vice President of Quantitative Analytics at Morgan Stanley, London (https://www.linkedin.com/in/harsh-prasad/)

Schedule

Preparation: Readings issued to all students at least 3 weeks before the start of the course. Students to complete all readings before the start of class.

Session 1 – Monday, June 10, 2018. 9 am – 12 pm. Instructor: Prof. Stephen Hansen.
Lecture on Latent Dirichlet Allocation (LDA), unsupervised learning, intro to graphical
models, illustrate output, etc. Python tutorial on Gibb’s sampling.

Session 2 – Tuesday, June 11, 2018. 9 am – 12 pm. Instructor: Prof. Stephen Hansen.
Bayesian Computation, Markov chain Monte Carlo (MCMC) and variational inference, with full algorithm derivation for LDA. Python tutorial.

Session 3 – Wednesday, June 12, 2018. 9 am – 12 pm. Instructor: Prof. Stephen Hansen.
Generative models for supervised learning (e.g. inverse regression, supervised LDA,
other ideas time permitting). Python tutorial.

Session 4 – Wednesday, June 12, 2018. 2 pm – 4 pm. Guest Speaker: Harsh Prasad.
Guest lecturer – Vice President from Morgan Stanley. Lecture on “Deep learning in Finance”.

Session 5 – Thursday, June 13. 9 am – 12 pm Instructor: Prof. Molly Roberts.
Structural topic modeling (Margaret E. Roberts, Brandon M. Stewart and Dustin Tingley).Causal inferences using text. R tutorial on Structural Topic Model.

Session 6 – Friday, June 14. 9 am – 12 pm: Instructor: Prof. Yaroslav Rosokha.
Overview of agent-based modelling for economics. Python tutorial.

Session 7 – Saturday, June 15. 9 am – 12 pm: Instructor: Prof. Yaroslav Rosokha.
Overview of reinforcement and evolutionary learning algorithms. Python tutorial.

Accommodation

We recommend the SwissTech Hotel right next to EPFL: https://www.swisstech-hotel.com/

Fees and discounts

We have a small budget to support accommodation expenses for each participant. Please reach out to the organizers for details (cmfe2019@epfl.ch) or message us on Facebook (click here).

Course fees:

  • PhD students: CHF 270
  • Master students: CHF 108

Register

If you are interested in the CMFE 2019 summer school, please register and pay here. Deadline: 20 May 2019.

Credits are only awarded to participants that paid the full course fee.