Machine Learners in Economics

Organizer: Dong Woo Hahm, Ph.D.

Date/Time: Mondays, 6 - 8 pm @KAP 319

Tentative meeting dates: sign up here.

• 2/24/2025 – Organizing Meeting

• 3/3/2025 – Regular Meeting

• 3/10/2025 – No meeting (Midterms)

• 3/17/2025 – No meeting (Spring Recess)

• 3/24/2025 – Regular Meeting

• 3/31/2025 – Regular Meeting

• 4/7/2025 – Regular Meeting

• 4/14/2025 – Regular Meeting

• 4/21/2025 – Regular Meeting

• 4/28/2025 – Regular Meeting

Format: 2 presenters per meeting, 1 hour for each presenter

• Presenters will present one of the following (please email me the slides in advance)

  • (i) One of the papers from the list below;

  • (ii) Any economics research paper (or reports) of their choice (please consult me before selecting);

  • (iii) Their own research project in ML and economics

• Suggested Structure:

  • 10 mins – Research Question & Background

    • What is the research question?

    • Why is this work important?

    • Brief context and motivation.

  • 20 mins – Overview of the Main ML Method

    • Explain the key ML technique used in the paper.

    • Focus on its intuition and how it applies to the research.

    • Think of this as an opportunity to learn yourself and teach the method to your peers (and me!).

  • 30 mins – Summary of Main Findings & Takeaways

    • What are the key results?

    • What are the policy or research implications?

    • Any limitations or open questions?

• Additional Notes:

  • No deep math required – prioritize intuition and logic over technical derivations.

  • This is NOT a class. There are no grades, no evaluations, and no reputational concerns. The goal is simply to learn together in a collaborative environment.

  • If you don’t fully understand something, that’s okay! Present what you do understand, and we’ll discuss and figure it out as a group.

Paper list:

1. "Machine Learning Methods for Demand Estimation" 

   • Link: https://www.aeaweb.org/articles?id=10.1257/aer.p20151021

2. "What Can Machines Learn, and What Does It Mean for Occupations and the Economy?"

   • Link: https://www.aeaweb.org/articles?id=10.1257/pandp.20181019

3. "A Machine Learning Approach to Analyze and Support Anti-corruption Policy"

   • Link: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3589545

4. "Augmenting Pre-Analysis Plans with Machine Learning"

   • Link: https://www.aeaweb.org/articles?id=10.1257/pandp.20191070

5. "Supervised Machine Learning for Eliciting Individual Demand"

   • Link: https://www.aeaweb.org/articles?id=10.1257/mic.20210069

6. "Machine Learning about Treatment Effect Heterogeneity: The Case of Household Energy Use"

   • Link: https://www.aeaweb.org/articles?id=10.1257/pandp.20211090

7. "Machine Learning in Economics and Finance"

   • Link: https://link.springer.com/article/10.1007/s10614-021-10094-w

8. "Machine Learning and Feature Selection: Applications in Economics and Climate Change"

   • Link: https://www.cambridge.org/core/journals/environmental-data-science/article/machine-learning-and-feature-selection-applications-in-economics-and-climate-change/4B729313EA2D42759A9C8E8A5C099D30

9. "The AI Economist: Optimal Economic Policy Design via Two-level Deep Reinforcement Learning"

   • Link: https://arxiv.org/abs/2108.02755

10. "Forecasting Four Business Cycle Phases Using Machine Learning: A Case Study of US and EuroZone"

    • Link: https://arxiv.org/abs/2405.17170

11. "Inference on Treatment Effects after Selection among High-Dimensional Controls"

    • Link: https://academic.oup.com/restud/article-abstract/81/2/608/1523757