Modelos Lineares Generalizados (2 º Sem 2011/2012)

Introduction to generalised linear models

1. Exponential family of distributions: introduction.

2. Natural and scale parameters. Mean and variance. Variance function.

3. Introduction to Generalized Linear Models: link functions, canonical link function, linear predictor.

4. Variables, factors, interactions. Parametrisation.

5. Deviance and scaled deviance. Model fit and model comparison.

6. Pearson and deviance residuals.

Statistical inference

1. Test of hypotheses on individual parameters.

2. Test of linear restrictions - nested models.

3. Estimation of dispersion parameter.

Continuous response models

1. The Normal model.

2. The Exponential and Gamma models.

Discrete response models

1. The Binomial model.

2. The Poisson model.

3. Modeling of proportions.

Quasi-likelihood and overdispersion

1. Introduction to quasi-likelihood estimation.

2. Likelihood equations for the general and regression models.

3. Choice of mean value and variance functions.

4. Estimation of the dispersion parameter.