Modelos Lineares Generalizados e Aplicações

Objectives

To understand and to be able to use binary, Poisson, gamma, log-linear regression models, framing them into the vaster class of models, the generalized linear models.

To know each of the models assumptions and to be able to say that they are met in practice.

To estimate the models using a specific software (R-project), to know how to interpret the estimates that are obtained and to assess the quality of the fits.

To study discrepant and influential observations and know how to deal with them.

To analyze several practical examples, mainly real, using the proposed methodologies.

General characterization

Code

12018

Credits

6.0

Responsible teacher

Inês Jorge da Silva Sequeira

Hours

Weekly - Available soon

Total - Available soon

Teaching language

Português

Prerequisites

Knowledge of Simple and Multiple Linear Regression Analysis.

 

Bibliography

Cordeiro GM, Demétrio CGB. (2008). Modelos Lineares Generalizados e Extensões. http://www.lce.esalq.usp.br/arquivos/aulas/2010/LCE5868/livro.pdf

Dobson AJ (1990). An Introduction to Generalized Linear Models. Chapman & Hall/CRC

Venables WN, Smith DM, R Core Team (2014). An Introduction to R: Notes on R: A Programming Environment for Data Analysis and Graphics, Version 3.1.0. http://www.r-project.org/

Hosmer DW, Lemeshow S (2013). Applied Logistic Regression – 3rd Edition. Wiley

McCullagh P, Nelder JA (1989). Generalized Linear Models - 2nd Edition. Chapman & Hall/CRC

Rodríguez, G. (2007). Lecture Notes on Generalized Linear Models. http://data.princeton.edu/wws509/notes/

Turkman MAA, Silva GL (2000). Modelos Lineares Generalizados - da teoria à prática. SPE http://docentes.deio.fc.ul.pt/maturkman/mlg.pdf

Teaching method

Available soon

Evaluation method

Two moments of evaluation:  individual report (50%) and final test (50%).

Subject matter

UA 1 – Linear Regression Model 

UA 2 – Generalized Linear Models  – definition, estimation, residuals, fit quality assessment, discrepant and influent observations

UA 3 – Gamma regression model and appications

UA 4 – Binary regression model (logistic) and applications 

UA 5 – Poisson regression model and applications

UA 6 – Log-linear model for two-dimensional tables and applications 

Programs

Programs where the course is taught: