The course covers new approaches in the areas of statistical modeling and data analysis, using ideas that bridge the gap between statistics and computer science and developing tools for the statistical mining of big data. The focus is on predictive learning, with particular emphasis on recent advances in regression and classification.

Program:

1) Introduction to statistical learning.

2) Dimension reduction. Principal Component Analysis.

3) Linear Models. Simple and multiple linear regression. Estimating the coefficients, assessing the accuracy of the coefficient estimates, assessing the accuracy of the model. Qualitative predictors. Model selection and regularization: subset selection, shrinkage methods (ridge regression and lasso), dimension reduction methods.

4) Supervised classification. Logistic regression. Linear and Quadratic discriminant analysis.

5) Unsupervised classification. Hierarchical clustering, K-means clustering.

6) Resampling methods. Cross-validation. The bootstrap.

7) Tree-based methods. Classification and regression trees. Bagging, random forests, boosting.

Following a blended learning approach, the course will make extensive use of the Statistical Learning MOOC by Hastie and Tibshirani referenced in the Bibliography. All methods will be illustrated using applications from marketing, finance, biology and other areas; the R free software environment for statistical computing and graphics (downloadable at www.r-project.org ) will be used and illustrated throughout the course and its lab sessions.

Through the course, students are expected to work in team on a real data analysis project whose progress will be shown periodically to the class.

The exam consists of two parts: a written exam and the presentation of the data analysis team project. To obtain a positive final mark for the course the student should pass each part of the exam with a grade greater than or equal to 18/30; the final mark is then obtained as the weighted average of two grades, with weights respectively equal to 0.6 for the written exam and 0.4 for the project presentation