Corso di studio in Economia

The course introduces to the fundamental techniques of statistical learning aimed at building a model for predicting a response variable based on one or more independent variables (or covariates). Special attention will be devoted to computer-based implementation of such techniques using a statistical software and to the interpretation of the analyses' results.

- Knowledge and understanding
The student will learn the most common methodologies for analyzing a data set together with their implementation through the software R. The student will also be able to interpret the results of the analysis and present them through both visual and numerical summaries.

- Applying knowledge and understanding
The student will have the ability to discuss various methods and techniques for statistical learning.
- Making judgements
The student will be able to select the appropriate statistical method for analyzing a datasets with the support the R software in supervised learning.
- Communication skills.
Students will properly use statistical language to comunicate the results of their findings.

Roughly half of the course will be delivered through lectures and half through labs for model implementation.

The final examination consists in a written test. Homeworks may be assigned during the course, whose evaluation may contribute to the final grade.

Statistical learning
- Goals
- Accuracy vs. interpretability
- Bias-variance trade off
Linear regression
- Simple linear regression
- Multiple linear regression
- Discussion and comparisons
Validation and resampling
- Cross-validation
- The bootstrap
Model selection and regularization
- Subset selection
- Shrinkage methods (ridge, lasso)
- Dimension reduction
Non-linear models
- Logistic regression for binary dependent variables
- Polynomial regression
- Regression Splines
- Generalized additive models

Testi adottati
James, Witten, Hastie, Tibshirani. An Introduction to Statistical Learning. Springer (2013)
Testi per approfondimento
Hastie, Tibshirani, Friedman. The Elements of Statistical Learning. Springer (2009)