Missing Data: Introducing the Multiple Imputation Doctor 
Introducing a decision making system for multiple imputation (midoc) & launch of 'Multiple Imputation and its Application' (2nd edition)
Join the Centre for Data and Statistical Science for Health for this interactive workshop and book launch on the topic of multiple imputation
Missing data is a common issue in health and social research, often addressed by multiple imputation (MI). MI is a flexible and general approach, with a suite of software packages. However, using MI in practice can be complex. Application of MI involves multiple decisions which are rarely justified or even documented, and for which little guidance is available. 
The Multiple Imputation DOCtor (midoc) R package is a decision-making system which incorporates expert, up-to-date guidance to help you choose the most appropriate analysis method when there are missing data.  
In this interactive workshop, you will have the opportunity to explore the key features of midoc using a worked example. With guidance from the package authors, attendees will learn how to use midoc to examine both the hypothesised causal relationships and the observed data to advise on whether MI is needed, and if so how to perform it. 
Workshop prerequisites:
- Each attendee will need a laptop with version 4.3.1 or later of R installed. Before the day of the workshop, attendees should install the midoc R package from GitHub using the following command: remotes::install_github("elliecurnow/midoc", build_vignettes=TRUE).
- Please note, the software is undergoing rapid development, so please re-install the latest version in the week before the workshop. In the event of any problems with installation, please contact Ellie Curnow.
- Some familiarity with missing data methods is essential; in addition, some knowledge of directed acyclic graphs (DAGs) and R is helpful.
- If you want to apply midoc to your own research question during the workshop, you will get the most out of it if you come with a DAG/dataset in mind.
The workshop will be followed by a book launch: Multiple Imputation and its Application (2nd edition) by James Carpenter, Jonathan Bartlett, Tim Morris, Angela Wood, Matteo Quartagno, and Michael Kenward.
This comprehensively revised second edition, includes an overview of the issues raised by missing data, the rationale for multiple imputation as a solution, and the practicalities of applying it in a multitude of settings. 
New material includes the use of multiple imputation for causal models, prediction models, measurement error and survey data; sections on imputation with non-linear relationships, multilevel models and sensitivity analyses have been revised and extended. 
A concluding chapter answers frequently asked multiple imputation questions, and gives a practical framework for applications. 
The book includes a wide range of theoretical and computer-based exercises, tested in the classroom, which are especially useful for users of R or Stata. Solutions are available on the companion website. 
±Ê°ù´Ç²µ°ù²¹³¾³¾±ð 
15.00-16.15: Introducing the Multiple Imputation Doctor (midoc) workshop- a decision-making system for multiple imputation (LG24)
16.15-16.30: Refreshments (LG24)
16.30-17.00: Book launch (Manson Lecture Theatre)
Speakers
Ellie Curnow PhD, University of Bristol 
Ellie Curnow conducts research into missing data methods as part of an MRC Better Methods, Better Research-funded project. After completing a PhD in medical statistics at the University of Bristol in missing data methods for survival analysis, she has built on her detailed knowledge of analysis in the presence of missing data to resolve questions around bias due to model misspecification when performing multiple imputation. As part of this, she has developed the midoc R package to guide researchers when performing analyses with missing data.  
James R. Carpenter, LSHTM & UCL
James R. Carpenter is Professor of Medical Statistics at the London School of Hygiene & Tropical Medicine and Programme Leader in Methodology at the MRC Clinical Trials Unit at UCL, UK.
Event notices
- Please note that you can join this event in person or you can join the session remotely (although one-to-one help with the midoc package will only be available to those attending in person)
- Please note that the recording link will be listed on this page when available.
Admission
Contact
