useR!2017 has ended
Back To Schedule
Friday, July 7 • 11:36am - 11:54am
Generating Missing Values for Simulation Purposes: A Multivariate Amputation Procedure

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

Feedback form is now closed.
Keywords: R-function ampute, Multivariate Amputation, Missing Data Methodology, Simulation Studies
Webpages: https://github.com/RianneSchouten/mice/blob/ampute/vignettes/Vignette_Ampute.pdf, https://www.rdocumentation.org/packages/mice/versions/2.30/topics/ampute, https://cran.r-project.org/web/packages/mice/index.html
Abstract: Missing data are a ubiquitous problem in scientific research, especially since most statistical analyses require complete data. To evaluate the performance of methods dealing with missing data, researchers perform simulation studies. An important aspect of these studies is the generation of missing values in complete data (i.e. the amputation procedure) and this procedure will be our focus.
Since no amputation software was available, we developed and implemented an extensive amputation procedure into an R-function: ampute (available in multiple imputation package mice). We will show that the multivariate amputation approach generates legitimate missing data problems.
We will provide evidence that ampute overcomes the problems of stepwise univariate amputation. With ampute, we have an efficient amputation method to accurately evaluate missing data methodology.


Friday July 7, 2017 11:36am - 11:54am CEST
2.01 Wild Gallery