Introduction to Machine Learning in Agricultural Economics, 3.5 ECTS

Prerequisites:

Ongoing PhD studies in social sciences/business studies/economics (interested students in related fields are subject to agreement). Students should have basic knowledge of programming language (R/ Python). No prior knowledge of ML required.

Objectives:

After successfully participating in the course, students will be able to :

• Have a general understanding of the possibilities and limitations of Machine Learning and understand core principles as well as difference between ML and Econometrics.  
• Have a theoretical and applied knowledge of common ML algorithms.
• Recognize relevant areas of application and understand differences in applicability across algorithms (No free lunch theorem) .
• Understand key evaluation methods and critically assess application cases and outcomes.
• Use and apply structured and unstructured data sources, identify relevant data sources.

Content:

• Key methods in ML and their application: supervised, unsupervised, and deep learning.
• Practical applications and recent advances for causal and predictive empirical research.
• Data analytics of heterogenous sources of data.

Examination:

• Students actively participate in the course and contribute to discussions during the course.
• Students conceptualize an application case and prepare a short presentation that they will pitch at the end of the course.
• Students write a course paper that applies methods and good practices introduced in the course.
• Students submit the script of the data analysis underling the course paper.

Preparation :

• Familiarize yourself with R/Python.
• Read the papers and the introduction chapters of the books of the literature list.
• (To be adjusted)

Literature :

• Athey, S., & Imbens, G. W. (2019). Machine learning methods that economists should know about. Annual Review of Economics, 11(1), 685–725. 
• Cui, P., & Athey, S. (2022). Stable learning establishes some common ground between causal inference and machine learning. Nature Machine Intelligence, 4(2), 110–115. https://doi.org/10.1038/s42256-022-00445-z 
• Lesmeister, C. (2019). Mastering machine learning with R: advanced machine learning techniques for building smart applications with R 3.5. Packt Publishing Ltd. 
• Storm, H., Baylis, K., & Heckelei, T. (2020). Machine learning in agricultural and applied economics. European Review of Agricultural Economics, 47(3), 849–892. 
• Swamynathan, M. (2017). Mastering machine learning with python in six steps: A practical implementation guide to predictive data analytics using python. Springer. 
• Additional literature will be provided before and during the course.
  
  

Additional information:

This course is part of the research school People, Society and Sustainability, a joined research school between the Department of Economics and the Department of Urban and Rural Development. Link to syllabus: Introduction to Machine Learning in Agricultural Economics (P000171)

Contact for registration and further information:

On course content, dates and examination: Lisa Höschle, lisa.hoschle@slu.se
On overarching organizational questions: Jens Rommel, jens.rommel@slu.se   

June 1 – Foundations & Applications  

09 – 12: Key principles, limitations and opportunities, terminology, comparison to econometrics, ML in the context of AI  

13 – 16: Current applications of ML (in AgEcon), Exercises in R/ Python,  

June 2 – Supervised Learning   

09 – 12: Classification, Regression, and Model Selection  

Linear & Non-linear Models, Feature Selection & Regularization, Model Diagnostics & Tuning, Model Explainability   

13 – 16: Exercise in R/ Python    

June 3 – Unsupervised Learning  

09 – 12: Clustering, Dimensionality Reduction (f.e in Big Data)  

13 – 16: Exercise in R/ Python  

June 4 – Deep Learning   

09 – 12: Neural Networks, Time Series & Forecasting  

13 – 16: Text mining/ Causal ML  

June 5 – Causal ML   

09 – 12: Combining ML & Econometrics, Policy Evaluation  

13 – 16: Student presentations: Pitch Course Paper, Feedback Round