MSA650 Linear mixed models for longitudinal data 7,5 hec

This course is an introduction to the area of mixed models which has become a necessary tool for treating real life situations with e.g. random effects, correlated observations and missing data. The emphasis is on longitudinal data and on how to use SAS and R to analyse mixed models. The course deals with the following topics:
  • Exploratory Data Analysis,
  • Estimation of the Marginal Model, Inference for the Marginal Model,
  • Inference for the Random Effects,
  • Fitting Linear Mixed Models with SAS, General Guidelines for Model Building,
  • Exploring Serial Correlation, Local Influence for the Linear Mixed Model ,
  • The Heterogeneity Model, Conditional Linear Mixed Models,
  • Exploring Incomplete Data, Joint Modeling of Measurements and Missingness,
  • Simple Missing Data Methods, Selection Models,
  • Pattern-Mixture Models, Sensitivity Analysis for Selection Models,
  • The Expectation-Maximization Algorithm, Design Considerations, Case Studies.
Syllabus
 
The course is
  • given every second year
  • in the first half of spring
  • jointly with Chalmers MVE210
 

Course information 2019

Course information 2017

  • Course coordinator: Ziad Taib
  • Schedule 2017

Course information 2015

  • Course coordinator: Ziad Taib
  • Schedule

Course information 2013

  • Course coordinator: Ziad Taib
  • Schedule

Course information 2011

  • Course coordinator: Ziad Taib
  • Schedule

Course information 2009

  • Course coordinator: Ziad Taib
  • Schedule

Published: Tue 18 Feb 2020.