Advanced Multiple Imputation Methods to Deal with Missing Data
Multiple imputation is a principled approach to account for missing data in analyses where valid results depends on careful construction of the imputation model. The potential for misspecification of the imputation model depends on several factors including the complexity of the analysis of interest, assumed reasons for missing data and the types of variables to be imputed. This course will introduce you to advanced multiple imputation methods that have been developed to address complex missing data analyses.
| Date | 11 December 2023 |
|---|---|
| Fee | £220 |
| Format | Online |
| Audience | Open to all applicants (prerequisites apply) |
Course profile
This course builds on prior knowledge of multiple imputation for dealing with missing data, and extends this to the application of multiple imputation in complex analyses.
Please click on the sections below for more information.
Structure
This 1-day course will be online and consist of pre-recorded lectures, live summary plus question & answer sessions, and live computer practical exercises. Each pre-recorded lecture will be followed by a live session which will summarise the main points of the lecture and allow participants to ask questions about the contents of the lecture.
In the afternoon, two topics will be run concurrently: (topic A) multiple imputation when the substantive analysis is a multilevel model, and (topic B) multiple imputation when the substantive analysis is a survival model. Participants will only be able to attend the live sessions of one topic (i.e., either A or B). However, after the course, participants will have access to the following on both topics A and B: pre-recorded lectures, computer practical materials, recordings of the live summary and question and answer sessions, and recordings of the computer practical demonstration.
The course is timetabled to start at 08:30 and end at 17:30, including time for coffee breaks and lunch.
Intended Learning Objectives
By the end of the course participants should be able to understand and use advanced multiple imputation methods when:
- data are missing not at random;
- the substantive analysis of interest is a multilevel model;
- the substantive analysis of interest is a survival model with partially observed covariates; and
- the substantive analysis of interest is a propensity score analysis with partially observed covariates.
Target audience
This course is intended for statisticians, health economists, epidemiologists and other researchers who are involved in performing statistical analyses of epidemiological datasets with missing data.
It is assumed that participants will have attended the Multiple Imputation for Missing Data course (or an equivalent introductory course to missing data concepts and multiple imputation). Participants should be familiar with: the concept of multiple imputation and have used it in practice; standard regression methods for dichotomous and continuous outcomes beyond the basic introductory level; and using software packages, Stata or R, for statistical analyses of the data.
Outline
The course will include:
- brief revision of theory and practice of multiple imputation in simple scenarios;
- guidance on conducting analyses when data are missing not at random;
- introduction to multiple imputation of covariates of a survival analysis model;
- introduction to multilevel multiple imputation;
- guidance on using R package jomo to impute multilevel data; and
- recommendations on the optimal way to implement multiple imputation for a propensity score analysis.
Teaching staff
Teaching staff will be drawn from researchers at the Department of Population Health Sciences and MRC IEU, who develop, test and provide guidelines on cutting edge missing data methodology. Course organisers, Dr Rachael Hughes and Dr Paul Madley-Dowd, are active members of this missing data methodology group.
Prerequisites
To make sure the course is suitable for you and you will benefit from attending, please ensure you meet the following prerequisites before booking:
| Knowledge |
Prior attendance of the Multiple Imputation for Missing Data short course (or equivalent introductory course to missing data concepts and multiple imputation) or be familiar with the concept of multiple imputation, and have used it in practice. Also, familiarity with standard regression methods for continuous and binary outcomes beyond a basic level. |
|---|---|
| Software |
You must have Stata* (version 13 or later) or R** (version 4.0.3 or later) installed in advance of the course. *Internal University of Bristol participants are given access to Stata. Go to Stata Installation Instructions (internal only) for help setting it up before the start of the course.
External participants are responsible for providing their own access to Stata, however if you are an employee of a university or another institution you may be able to get a short term free Evaluate license. If you are a student, Stata offer a short term free Student licence (one week).
**Go to R Installation Instructions for help getting set up.
|
| Recommendation | For the computer practicals, we recommend that participants either have access to the use of two screens or the ability to print materials in advance of the course. |
Bookings
Before booking this course, please make sure you read the information provided above about the target audience and prerequisites. It is important that you have access to the relevant IT resources needed for the course and meet the knowledge prerequisites to ensure you can get the most from the course.
Bookings are taken via our online booking system, for which you must register an account. To check if you are eligible for free or discounted courses please see our fees and voucher packs page. All bookings are subject to our terms & conditions, which can be read in full here.
For help and support with booking a course refer to our booking information page, FAQs or feel free to contact us directly. For available payment options please see: How to pay your short course fees.
Course materials
Participants are granted access to our virtual learning platform (Blackboard) 1 to 2 weeks in advance of the course. This allows time for any pre-course work to be completed and to familiarise with the platform.
To gain the most from the course, we recommend that you attend in full and participate in all interactive components. We endeavour to record all live lecture sessions and upload these to the online learning environment within 24 hours. This allows course participants to review these sessions at leisure and revisit them multiple times. Please note that we do not record breakout sessions.
All course participants retain access to the online learning materials and recordings for 3 months after the course.
University of Bristol staff and postgraduate students who do not wish to attend the full course may instead register for access to the 'Materials & Recordings' version of this course: Further information and bookings.
Testimonials
100% of attendees recommend this course*.
*Attendee feedback from 2023-2024.
Here is a sample of feedback from the last run of the course:
"The material (slides/ recordings and practical exercises) were all really clear and useful. I found the posit cloud really user friendly. It was good to have an option for demonstration for the practicals or work through it on your own with options to ask questions. I liked that the practicals were provided in both R and Stata (where possible). The course was well run" - course feedback, December 2023
"Amazing course!! a lot of tough concepts were probably explained and simplified by the tutors. Tutors were so easy to ask and collaborative" - course feedback, December 2023
"I think the Q&A were good and many helpful papers were supplied. The live practical go throughs were helpful" - course feedback, December 2023
"The resources provided are great. The lectures were well explained and at a good pace. Facilitators really helpful and provided lots of useful sources for peoples questions. Posit cloud was helpful to ensure fewer IT issues" - course feedback, December 2023
"Course delivery and the comprehensive responses to all raised questions, particularly the inclusion of supporting papers, were noteworthy" - course feedback, December 2023
"The course was very focused, with clear learning objectives and good materials. Rachel and Jon are experts in the field, willing to share their knowledge, which is wonderful for learners" - course feedback, December 2023
"It is the best course I have ever attended" - course feedback, December 2023
Bookings for this course are now closed
The course was very focused, with clear learning objectives and good materials. Rachel and Jon are experts in the field, willing to share their knowledge, which is wonderful for learners
Dates don't work? Just need a refresher?
Find out about the self-paced Materials & Recordings version of this course [UoB only].