Coursebooks 2018-2019

Modern regression methods

MATH-408

Lecturer(s) :

Davison Anthony C.

Language:

English

Remarque

Course given every two years (given in 2018-19)

Summary

A second course on regression modelling, dealing with nonlinear effects of explanatory variables, and non-normal and dependent response variables.

Content

Revision of linear regession and likelihood inference

 

Fitting algorithms for nonlinear models and related diagnostics

 

Generalised linear model; exponential families; variance and link functions

 

Proportion and binary responses; logistic regession

 

Count data and Poisson responses; log-linear models

 

Overdispersion and quasilikelihood; estimating functions

 

Mixed models, random effects, generalised additive models and penalized regression

 

Keywords

Binary response; Count data; Deviance; EM algorithm; Estimating function; Iterative weighted least squares algorithm; Lasso; Likelihood; Logistic regression; Longitudinal data; Mixed model; Multinomial distribution; Overdispersion; Poisson distribution; Quasi-likelihood; Random effects

 

Learning Prerequisites

Required courses

Knowledge of basic probability and statistics, at, for example, the levels of MATH-240 and MATH-230

 

Linear models (MATH-341) or equivalent

 

Important concepts to start the course

Linear regression; likelihood inference; R

 

Learning Outcomes

By the end of the course, the student must be able to:

• Develop theoretical elements needed in regression analysis
• Apply the statistical package R to the analysis of data
• Assess / Evaluate the quality of a model fitted to regression data, and suggest improvements
• Choose a suitable regression model

Transversal skills

• Demonstrate a capacity for creativity.
• Demonstrate the capacity for critical thinking
• Write a scientific or technical report.

Teaching methods

Ex cathedra lectures; homework both theoretical and practical; mini-project

 

Expected student activities

Attending lectures; solving theoretical problems; solving applied problems using statistical software

Assessment methods

Written final exam; mini-project

Dans le cas de l'art. 3 al. 5 du Règlement de section, l'enseignant décide de la forme de l'examen qu'il communique aux étudiants concernés.

Supervision

Office hours	Yes
Assistants	Yes
Forum	Yes

Resources

Bibliography

Davison, A. C. (2003) Statistical Models.  Cambridge University Press.

Ressources en bibliothèque

• Statistical Models / Davison