Reminder registration closes on 23rd March.
There are some small bursaries available for PhD students otherwise
unable to attend.
--------------------------------------------------------------------
Missing data is a common problem encountered by researchers from many
disciplines, including medical research, public health, criminology,
politics, biological and environmental science, psychology and the
applied social sciences to name but a few. This practical workshop will
help researchers deal more effectively with missing data.
Title: HANDLING MISSING DATA
Statistical modelling with missing data using multiple imputation and
inverse probability weighting
Presenter: Dr James Carpenter
Dates of Course: Monday 30th March - Wednesday 1st April 2009
Venue: Postgraduate Statistics Centre, Lancaster University (For more
info and costs see http://www.maths.lancs.ac.uk/missing-data)
Overview:
This short course will provide an introduction to the issues raised by
missing data and statistical modelling with missing data. In particular,
participants will gain an understanding of multiple imputation as a tool
for handling missing data.
They will learn how to implement this method through a series of
practical computing exercises with example datasets in Stata and MLwiN.
The course will have a strong practical focus with six computer sessions
to consolidate the ideas presented in the lectures, and to gain
experience with the various methods.
Target audience:
Social Scientists, epidemiologists, biostatisticians and other
researchers with strong quantitative skills and substantial experience
in statistical analysis including familiarity with multivariable
regression methods. During computing practical sessions the participants
will be provided with computing code, solutions and assistance.
It is strongly recommended that participants are familiar with Stata,
and to a lesser extent MLwiN.
More detail:
The main objectives of the course are:
* To introduce the key concepts underpinning the analysis of
partially observed data, together with a principled approach to the
analysis;
* To explain the shortcomings of frequently used ad-hoc methods
* To introduce mulitple imputation, and gain familiarity with using
the ICE software in Stata for multiple imputation,
and the MLwiN software for multiple imputation, using simple and
more complex examples
* To explore the role of sensitivity analysis, and methods for
performing approximate sensitivity analysis
* To introduce inverse probability weighting and doubly robust
estimation.
Course Content:
The course will consist of 6 sessions, each of which comprises a 1-h
lecture followed by a short discussion and then a 1.5h computer
practical. The key topics covered will be:
* Session I: Introduction, issues raised by missing data, and
towards a systematic approach
* Session II: Shortcomings of ad-hoc methods; introduction to
multiple imputation
* Session III: Further issues in multiple imputation
* Session IV: Multilevel multiple imputation
* Session V: Sensitivity analysis
* Session VI: Inverse probability weighting and doubly robust
estimation.
Computer workshops will enable course participants to put the methods
into practice. The course will use the packages Stata (80%) and Mlwin
(20%).
Course Materials:
Participants will receive written course notes.
The Instructor:
James Carpenter is a Reader in Medical and Social Statistics at the
London School of Hygiene & Tropical Medicine (University of London). His
main research interest is missing data, and developing and applying
multiple imputation. He has led missing data courses in the UK and
abroad, including under the ESRC's Researcher Development Initiative.
Preparatory Reading:
* Participants would benefit from reviewing the introductory
material on www.missingdata.org.uk (under Getting Started)
* See also Kenward, M. and Carpenter, J (2007): Muliple Imputation:
Current perspectives. Statistical Methods in Medical Research, 16,
199-218
|