Machine learning: Tools and applications for policy

During this course you will learn how to:

• Use popular Machine Learning (ML) and Artificial Intelligence (AI) techniques
• Understand the opportunities and limitations of these techniques
• Be able to interact with the experts
• Work hands-on with ML/AI methods in Python
• “Demystify” the black box of ML/AI
• Continue to learn by yourself

• Popular Machine Learning (ML) and Artificial Intelligence (AI) Techniques
• Opportunities and Limitations of ML and AI
• AI and ML methods in Python and practical exercises

• A master or PhD in any discipline that works with data. The main example used will be using credit risk data.
• Basic knowledge of traditional econometric methods (OLS, panel and time-series models) is assumed.
• The tutorials use Python running in a Jupyter Notebooks on Google Colab. This is browser based so no software installation need. A good internet connection is required. The material will also be distributed through Gitlab.
• Knowledge of Python is helpful but not is required.

Target Audience

Policy makers and analysts engaged in data-driven policy making.

We expect the participants to be familiar with basic data manipulation tasks (e.g. Excel) and have a working knowledge of linear regressions, that is to be able to run OLS in its variations and interpret the results.

The tutorials use Python in Jupyter Notebooks on Google Colab. This is browser based so no installation need. Knowledge of Python is helpful but not required. Basic coding skills and familiarity with Python is recommended. For a tutorial, see this here (link).