Flexible regression through Generalized Additive Models (GAMs). Biomedical applications using R
Professora Denisa Mendonça e Professor Pedro Oliveira


Carmen María Cadarso-Suárez

Unit of Biostatistics-School of Medicine.

University of Santiago de Compostela.





0.     Introduction to R.


1. Generalized Linear Models - GLMs

1.1. Some theory about the GLM.

1.2. Estimation and inference.

1.3. GLMs with R. Practical examples.


2. Introduction to flexible regression methods

2.1. Polynomial regression. Piecewise polynomial regression.

2.2. Regression splines: B-splines (bs), natural splines (ns).

2.3. Penalized regression splines.

2.4. Smoothing regression techniques.

2.4.1. Univariate smooth functions (Smoothers).

2.4.2. Smoothing bases: P-splines (ps),  thin plate regression splines (tp),..

2.4.3. Effective number of “degrees-of-freedom” of an smoother.

2.5. Exercises with R.


3. Generalized Additive Models– GAMs

3.1. Some theory about the GAM.

3.1.1. Estimation and inference.

3.1.2. Multivariate selection of smoothing parameters.

3.2. GAM including interactions. Tensor product bases (te).

3.3. GAMs with R.

3.3.1. The mgcv package.

3.3.2. GAMs in practice: Real data examples.


Hastie T.J. and Tibshirani T.J. (1990). Generalized Additive Models. Chapman & Hall.

McCullagh P. and Nelder J.A. (1989). Generalized Linear Models. Second Edition. Chapman &Hall/CRC.

Wood S. (2006). Generalized Additive Models. An introduction with R. Chapman & Hall/CRC.

Detailed Programme

Tuition Fee:
150 (ISPUP internal members); 200 (external members)
9 am - 18.30 pm
equivalente a 3 ECTS
Start date:
End date: