Course Details
Our aim in this course is to provide participants with the ability to analyse their own data using multiple imputation, but also to be aware of the pitfalls and limitations of the technique.
We will give plenty of practical examples from our own experience of analysing data in medical research.
We welcome participants bringing their own data and problems, and one session in day 2 is dedicated to "live" analysis of some participants' data.
A tutorial
based on this course has appeared in Statistics in Medicine.
Programme
(Timings are approximate)
Day 1
| 09:30 - 10:00 |
Registration, tea and coffee |
| 10:00 - 10:20 |
Introductions and announcements |
| 10:20 - 11:20 |
Lecture 1 Introduction to missing data and multiple imputation (Angela Wood) |
| 11:20 - 12:05 |
Practical 1 |
| 12:05 - 12:40 |
Lecture 2a Imputing a single incomplete variable (Patrick Royston) |
| 12:40 - 13:30 |
Lunch |
| 13:30 - 14:10 |
Lecture 2b Imputing multiple incomplete variables: introduction to chained equations (Tim Morris) |
| 14:10 - 15:00 |
Practical 2 |
| 15:00 - 15:30 |
Tea/coffee |
| 15:30 - 16:30 |
Lecture 3 Imputation by chained equations with non-Normal data (Patrick Royston) |
| 16:30 - 17:30 |
Practical 3 |
Day 2
| 09:15 - 10:15 |
Lecture 4 Specifying the imputation model, and some difficulties (Ian White) |
| 10:15 - 11:00 |
Practical 4 |
| 11:00 - 11:30 |
Tea/coffee |
| 11:30 - 12:00 |
Lecture 5 Model-building and post-estimation issues (Angela Wood) |
| 12:00 - 12:45 |
Practical 5 |
| 12:45 - 13:30 |
Lunch |
| 13:30 - 15:00 |
Open session Tackling imputation problems in participants' own data, and general discussion (all) |
| 15:00 - 15:55 |
Lecture 6 Pitfalls, limitations and extensions (Ian White) |
| 15:55 - 16:10 |
Tea/coffee |
| 16:10 - 16:40 |
Lecture 7 Planning and reporting analyses using multiple imputation (Angela Wood) |
|
![[BSU Logo]](/BSUsite/images/bsu_logo.gif){kind=logo}
![[Valid XHTML 1.1]](/BSUsite/graphics/valid-xhtml11.png)