Linear Mixed Models for Longitudinal Data, 7, 5 p (2015)

Lectures

 

 Teacher: Ziad Taib (0707655471 ) ziad.taib@astrazeneca.com or ziad@chalmers.se 

 

Textbook: Linear mixed models for longitudinal data, Geert Verbeke and Geert Molenberghs Springer Verlag, New York plus some handouts. The exact chapters that are due can be found below.

 

Schedule and Outline of Lectures: All lectures will take place in MVH11 Fridays 10.00-11.45 in the form of power point presentations. These will be handed out at the beginning of each lecture.

 

 

 March 24

Introduction to Linear Mixed Models (Chap 1-4)

 March 27

Estimation and Inference for the marginal model (Chap 5-6)

 April 24

Inference for the random effects, Software issues (Chap 7-8)

 May 8

Non-linear and Generalized Linear Mixed Models (Handouts)

 May 15

Incomplete data (Chap 15-16)

 May 22

Design and sample size issues (Chap 23)

 May 29

Model checking

 

 

Reading instructions:

 

Chapters 1-5

The whole chapter is required

Chapter 6

6.1 to 6.3.3

Chapter 7

7.1-7.7

Chapter 8

The whole chapter is required

Chapter 14

Self reading

Chapter 15

The whole chapter is required

Chapter 16

The whole chapter is required

Chapter 20

20.3

Chapter 23

The whole chapter is required

Handouts

Generalized mixed models

Handouts

Non-linear mixed models

Handouts

Model checking

 

 

Exam: The Final written exam will be a closed book exam where the necessary complicated formulas will be provided along with the examination text. No complicated calculations are required but you might need a simple pocket calculator. Both “theoretical” and applied problems can be posed. You should be able to interpret Proc Mixed results but you will not be required to write any complex code, SAS or otherwise. Out of a total of 30 (100%) possible scores, the computer assignments give a “bonus” of up to 6 scores (20%) while the written exam gives 24 scores (80%). There are three possible grades: Excellent (over 26 scores), Pass (over 16 scores) and do not pass (under 16 scores). Some sample questions covering parts of the course can be found here. Answers to these can be found here. A sample exam can be found here. The solutions to the exam can be found here.

 

Computer assignments

 

Teachers: Henrike Häbel (031 772 53 80) henrike.habel@chalmers.se and Alexandra Jauhiainen (031-7761000) alexandra.jauhiainen@astrazeneca.com for the computer assignments.

 

Assignments: There will be one pre-assignment to get started and 2 assignments that are part of the examination. These are to be solved and presented in a report, with programs, output, and statistical notation integrated. The quality of writing and organization, as well as content, will influence the grade. Homework may be turned in directly to Henrike Häbel. The deadlines for the assignments are May 12 for lab1 and May 26 for lab2. All computer lab sessions will take place in MVF22 31/3, 21/4, 28/4, 5/5, 12/5 and 19/5.

 

Computer access: All students are supposed to have a computer account. This account will enable computer lab use and use of computers in the School of Mathematical Sciences.

 

Software: This course will make extensive use of SAS. This software is available on the computers that will be used The application SAS SYSTEM ver 9.4 can be accessed by using a Remote Desktop Connection (RDP) to the computer 'winlab81.ita.chalmers.se'. A detailed description of how you can logon to this remote computer is available @ http://www.studat.chalmers.se/mofit/studat/rdp. Alternatively, R, Splus or Spss can be used to solve the home assignments but no instructions will be given in these programs.

 

 

 

Material for the computer assignment:

 

(i)          Pre-assignmentIntroduction to SAS – Data – SAS program.

(ii)         Assignment 1Lab 1 (.doc) – Lab1 (.pdf)- Data1 Zip – Data1 – Data2 – Data 3.

(iii)        Assignment 2Lab2 - Data.

 

News

 

(i)          Welcome to the course Linear Mixed Models on March 24 2015. Throughout the course, news on lectures, assignments etc. will be posted here.