A workshop on missing data with emphasis on sample selection models will take place in London in September. This workshop aims to provide an introduction to the issues raised by missing data with particular emphasis on missing not at random. The workshop is free but registration is required.
Who will benefit and how
This workshop provides a practical introduction to the use of a very flexible class of sample selection models which can be useful to analysts and quantitative researchers whose analyses are affected by missing not at random. There will be plenty of opportunity for discussion with the course instructors on sample selection models (but not only) for analysing missing data, and help will be provided with interpreting results. Throughout, there will be a focus on the policy implications of missing data and potential solutions to this problem. The workshop will also be relevant for practitioners conducting surveys on topics which are likely to be affected by missing information, as we will discuss how to incorporate methods to adjust for data not missing at random into the way surveys are designed.
Summary of workshop
This workshop aims to:
• Provide an introduction to the issues raised by missing data with particular emphasis on missing
not at random;
• Briefly introduce ad-hoc methods and principled methods;
• Briefly introduce pattern mixture models;
• Introduce and discuss sample selection models;
• Illustrate the use of SemiParBIVProbit to deal with missing not at random;
• Case studies from DHS studies, ageing studies and cost effectiveness analyses.
Workshop timetable
Thursday 15 September
10:00-10:15 Introduction (ALL)
10:15-11:00 Missing Mechanisms, Ad-Hoc and Principled Methods (GH)
11:00-11:30 Introduction to Pattern Mixture Models (MG)
11:30-11:45 Break
11:45-12:45 Sample Selection Models and Variants: Part I (GM/RR)
12:45-14:00 Lunch
14:00-14:45 Sample Selection Models and Variants: Part II (GM/RR)
14:45-15:15 Introduction to R Package SemiParBIVProbit (GM/RR)
15:15-15:30 Break
15:30-16:30 Practicals with SemiParBIVProbit (ALL)
Friday 16 September
10:00-10:30 Missing data in ageing studies (MM)
10:30-11:00 Missing data in DHS studies (or South Africa work) (GH)
11:00-11:15 Break
11:15-11:45 Missing data in cost-effectiveness analyses (MG)
11:45-12:15 Designing survey to incorporate sample selection approaches (MM)
12:15-13:00 Interactive discussion (ALL)
Prerequisites
Participants are expected to have some familiarity with R, as well as relevant statistical concepts such as linear regression.
Registration
Who will benefit and how
This workshop provides a practical introduction to the use of a very flexible class of sample selection models which can be useful to analysts and quantitative researchers whose analyses are affected by missing not at random. There will be plenty of opportunity for discussion with the course instructors on sample selection models (but not only) for analysing missing data, and help will be provided with interpreting results. Throughout, there will be a focus on the policy implications of missing data and potential solutions to this problem. The workshop will also be relevant for practitioners conducting surveys on topics which are likely to be affected by missing information, as we will discuss how to incorporate methods to adjust for data not missing at random into the way surveys are designed.
Summary of workshop
This workshop aims to:
• Provide an introduction to the issues raised by missing data with particular emphasis on missing
not at random;
• Briefly introduce ad-hoc methods and principled methods;
• Briefly introduce pattern mixture models;
• Introduce and discuss sample selection models;
• Illustrate the use of SemiParBIVProbit to deal with missing not at random;
• Case studies from DHS studies, ageing studies and cost effectiveness analyses.
Workshop timetable
Thursday 15 September
10:00-10:15 Introduction (ALL)
10:15-11:00 Missing Mechanisms, Ad-Hoc and Principled Methods (GH)
11:00-11:30 Introduction to Pattern Mixture Models (MG)
11:30-11:45 Break
11:45-12:45 Sample Selection Models and Variants: Part I (GM/RR)
12:45-14:00 Lunch
14:00-14:45 Sample Selection Models and Variants: Part II (GM/RR)
14:45-15:15 Introduction to R Package SemiParBIVProbit (GM/RR)
15:15-15:30 Break
15:30-16:30 Practicals with SemiParBIVProbit (ALL)
Friday 16 September
10:00-10:30 Missing data in ageing studies (MM)
10:30-11:00 Missing data in DHS studies (or South Africa work) (GH)
11:00-11:15 Break
11:15-11:45 Missing data in cost-effectiveness analyses (MG)
11:45-12:15 Designing survey to incorporate sample selection approaches (MM)
12:15-13:00 Interactive discussion (ALL)
Prerequisites
Participants are expected to have some familiarity with R, as well as relevant statistical concepts such as linear regression.
Registration
The event is free but prior registration is required. To reserve a place, please contact Rosalba Radice r.radice@bbk.ac.uk and provide a brief explanation of your interest in the course and short CV.
Instructors
Rosalba Radice (Birkbeck, Department of Economics, Mathematics and Statistics)
http://www.ems.bbk.ac.uk/faculty/radice
Giampiero Marra (University College London, Department of Statistical Science)
http://www.ucl.ac.uk/statistics/people/giampieromarra
Manuel Gomes (London School of Hygiene and Tropical Medicine, Department of Health Services Research and Policy)
http://www.lshtm.ac.uk/aboutus/people/gomes.manuel
Guy Harling (Harvard School of Public Health, Department of Global Health and Population)
https://www.hsph.harvard.edu/guy-harling/
Mark McGovern (CHaRMS, Queen's University Belfast)
http://go.qub.ac.uk/MarkMcGovern
Key references
1. Marra, G., Radice, R., Bärnighausen, T., Wood, S.W., McGovern, M.E. (in press). A Simultaneous Equation Approach to Estimating HIV Prevalence with Non-Ignorable Missing Responses. Journal of the American Statistical Association
2. Marra, G., Radice, R. (2016). A Bivariate Copula Additive Model for Location, Scale and Shape.
3. Marra, G., Radice, R. (2016). SemiParBIVProbit: Semiparametric Copula Bivariate Probit Modelling. R package version 3.7-1.
4. McGovern, M.E., Marra, G., Radice, R., Canning, D., Newell, M.L., Bärnighausen, T. (2015). Adjusting for Non-Participation Bias at an HIV Surveillance Site in Rural South Africa. Journal of the International AIDS Society, 18, 19954.
5. Marra G., Radice R. (2013). A Penalized Likelihood Estimation Approach to Semiparametric Sample Selection Binary Response Modeling. Electronic Journal of Statistics, 7, 1432-1455.
6. Marra G., Radice R. (2013). Estimation of a Regression Spline Sample Selection Model. Computational Statistics and Data Analysis, 61, 158-173.
7. Radice R., Marra G., Wojtys M. (in press). Copula Regression Spline Models for Binary Outcomes. Statistics and Computing.
8. McGovern M.E., Bärnighausen T., Marra G., Radice R. (2015). On the Assumption of Joint Normality in Selection Models: A Copula Approach Applied to Estimating HIV Prevalence. Epidemiology, 26(2), 229-237.
9. Bärnighausen, T., Bor, J., Wandira-Kazibwe, S., Canning, D. (2011). Correcting HIV prevalence estimates for survey nonparticipation using Heckman-type selection models. Epidemiology, 22, 2735.
7. Radice R., Marra G., Wojtys M. (in press). Copula Regression Spline Models for Binary Outcomes. Statistics and Computing.
8. McGovern M.E., Bärnighausen T., Marra G., Radice R. (2015). On the Assumption of Joint Normality in Selection Models: A Copula Approach Applied to Estimating HIV Prevalence. Epidemiology, 26(2), 229-237.
9. Bärnighausen, T., Bor, J., Wandira-Kazibwe, S., Canning, D. (2011). Correcting HIV prevalence estimates for survey nonparticipation using Heckman-type selection models. Epidemiology, 22, 2735.
No comments:
Post a Comment