A GUI for Missing Data Exploration

Provides numeric and graphical summaries for the missing values from both categorical and quantitative variables. A variety of imputation methods are applied, including the univariate imputations like fixed or random values, multivariate imputations like the nearest neighbors and multiple imputations, and imputations conditioned on a categorical variable.


GTK+ is required to install this package. To install GTK+, try the following line in R:


If any error occurs, then run the above line again. If GTK+ can still not be installed, visit http://www.gtk.org/download/ and install GTK+ 2.x. (For Mac OS users, visit http://r.research.att.com/ to get the recent GTK+ 2.x framework binary.) Then run the above R code again.

After installing the RGtk2 package, run



Changes in Version 0.2-5

  • Fix the color bugs in pairwise plot and parallel coordinates plot

Changes in Version 0.2-4

  • Small changes based on the new version of package mi
  • Add citation
  • Update the affiliation information

Changes in Version 0.2-3

  • Small changes based on the new version of package mi

Changes in Version 0.2-2

  • In the 'Simple' method, 'Mean/Mode' is renamed by 'Mean'. Mode applies when median or mean does not work.
  • Update the parameter for hclust (from 'ward' to 'ward.D')
  • mi from the mi package only accepts n.imp>1, so we set n.imp=2 when the actual input is 1, and take the first set as the result.

Changes in Version 0.2-0


  • Add new imputation methods (multiple imputations and nearest neighbors), re-arrange the widget for methods.

  • Involve the tabs for the graphic device, which can display different methods or multiple chains in parallel.

  • Add a new graph for parallel coordinates plot, which sorts the variables by F-statistics of ANOVA.

  • Add a "Setting" tab to tune the parameters.

  • Fix the bugs in the graphical device for Windows platform.

Changes in Version 0.1-6

Edit NAMESPACE by required.

Changes in Version 0.1-4


  • A new window is created when hitting the button "Export Data" on the main GUI. This window provides the following options for exporting the imputed data:

    • Whether to export the data of all the variables or only the selected variables. The default setting is all variables. If the user does not select anything when pressing the "Export Data" button, then all the variables are highlighted/selected.

    • Whether to export the shadow matrix. Shadow matrix is useful to indicate the places of missing values / imputed values after the imputed data replace NA's.

    • Where to save the exported data, i.e. to choose the directory.

    • What name to use for the exported data.

    • What format to save the data. Users can save them in a csv format or rda format file, or export the data to R console directly as a data frame.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.2-5 by Xiaoyue Cheng, 3 years ago

Browse source code at https://github.com/cran/MissingDataGUI

Authors: Xiaoyue Cheng , Dianne Cook , Heike Hofmann

Documentation:   PDF Manual  

GPL (>= 2.0) license

Imports GGally, cairoDevice, grid, reshape

Depends on gWidgetsRGtk2, ggplot2

Suggests Hmisc, norm, mice, mi

See at CRAN