90352 - Machine Learning For Economists

Learning outcomes

At the end of the course the student will have a good understanding of the main tools used in machine learning. In particular, he/she: - understands and knows how to apply key aspects of machine and statistical learning, such as out-of-sample cross-validation, regularization and scalability - is familiar with the concepts of supervised learning, regression and classification - understands and can apply the main learning tools such as lasso and ridge regression, regression trees, boosting, bagging and random forests, support vector machines and neural nets. - The course will put special emphasis on empirical applications using the R software.

Course contents

1. Introduction and Overview of Statistical Learning
   
2. Resampling Methods: Cross-Validation and the Bootstrap
3. Linear Model Selection and Regularization: Ridge Regression, the LASSO and Principal Components
4. Moving Beyond Linearity: Regression Splines, Smoothing Splines and General Additive Models
5. Tree-based Methods and Neural Networks
6. Unsupervised Learning: Dynamic Factor models, Networks

Readings/Bibliography

- G. James, D. Witten, T. Hastie, R. Tibshirani, An Introduction to Statistical Learning with Applications in R, 2nd Edition, 2021

- T. Hastie, R. Tibshirani, J. Friedman, The Elements of Statistical Learning, 2nd Edition, 2009

- Lecture notes

- Selected papers

Teaching methods

For each topic we will first introduce the relevant theory and the most relevant mathematical aspects, and then move to its empirical application in the R language or Matlab (for factor models). Special emphasis will be placed on the economic interpretation of the results. Knowledge of time series as taught in Macroeconometrics is strongly recommended.

Assessment methods

The exam consists in writing a short paper on a project assigned by the teacher and related to the topics covered during classes.

Alternatively, the student can propose a topic of interest whose consistency with the course contents must be evaluated and approved by the teacher.

In both cases the students are required to give an oral presentation of their work.

Depending on the chosen project students can work in groups formed of maximum 4 people.

The maximum possible score is 30 e lode. The exam is graded as follows:

<18 failed
18-23 sufficient
24-27 good
28-30 very good
30 e lode excellent

The final grade can be rejected only once.

 

Teaching tools

Slides, lecture notes for the theory part.

R packages and Matlab codes to discuss empirical analysis and replicate the results of a few papers.

Office hours

See the website of Matteo Barigozzi