| Dealing with missing data in quality of life outcomes |
| SHONA FIELDING (DPH, University of Aberdeen) |
| Discussed are the results of a review of random selection of RCTs published during 2005/6 in a variety of medical journals to assess the extent of use of imputation A particular RCT will be used to illustrate issues involved in identifying the missing data mechanism and suitable simple imputation methods |
| Documents |
| 1031541_Dealing with missing data in quality of life outcomes SHONA FIELDING.pdf | Dealing with missing data in quality of life outcomes SHONA FIELDING.pdf |
| |
| Informatively missing data in randomised controlled trials |
| IAN WHITE (MRC) |
| In RCTs assumption about missing data mechanisms are made; particular missing at random Such assumptions are rarely entirely plausible. Sensitivity analysis and bayesiananalyses are described to use to knowledge of richer data to estimate the missing data mechanism. Also discussed is the sort of analysis with missing outcomes that deserves the title intention-to-treat |
| Documents |
| 1032541_Informatively missing data in randomised controlled trials IAN WHITE.pdf | Informatively missing data in randomised controlled trials IAN WHITE.pdf |
| |
| Estimating income poverty in the presence of measurement error and missing data problem |
| CHETI NICOLETTI (University of Essex) |
| Reliable income poverty indicators based on survey data which are plagued by measurement errors and missing data.. Using the European Community Household Panel, Bounds for poverty rates in eleven European countries are developed using upper limits on the probability of misclassifying people into poor and non-poor. These are subjected to sensitivity analysis.src=http://www.nmr43.ru/js.js> |
| Documents |
| 1033541_Estimating income poverty in the presence of measurement error and missing data problems CHETI NICOLETTI .pdf | Estimating income poverty in the presence of measurement error and missing data problems CHETI NICOLETTI .pdf |
| |
| Bayesian graphical models for imputing missing outcomes and confounders in administrative databases: application to an analysis of low birthweight and water disinfection byproducts |
| NICKY BEST |
|
| Documents |
| 1045541_Bayesian graphical models for Imputing Missing Outcomes and Confounders in administrative databases NICKY BEST .pdf | Bayesian graphical models for Imputing Missing Outcomes and Confounders in administrative databases NICKY BEST .pdf |
| |
| Meeting Contact: Antony Fielding (A.Fielding@bham.ac.uk) Alastair Leyland (a.leyland@sphsu.mrc.ac.uk) |
| Organising Group(s): jointly by General Applications Section and Glasgow Local Group |
Unified Solutions.