Graphical user interface for visualization and imputation of missing values.
The Data menu allows to select a data set from the workspace or load
data into the workspace from
RData files. Furthermore, it can be
used to transform variables, which are then appended to the data set in use.
Commonly used transformations in official statistics are available, e.g.,
the Box-Cox transformation and the log-transformation as an important
special case of the Box-Cox transformation. In addition, several other
transformations that are frequently used for compositional data are
implemented. Background maps and coordinates for spatial data can be
selected in the data menu as well.
After a data set was chosen, variables can be selected in the main menu, along with a method for scaling. An important feature is that the variables will be used in the same order as they were selected, which is especially useful for parallel coordinate plots. Variables for highlighting are distinguished from the plot variables and can be selected separately. For more than one variable chosen for highlighting, it is possible to select whether observations with missing values in any or in all of these variables should be highlighted.
A plot method can be selected from the Visualization menu. Note that plots that are not applicable to the selected variables are disabled, for example, if only one plot variable is selected, multivariate plots cannot be chosen.
The Imputation menu offers robust imputation methods to impute variables of the data set.
The Diagnostics menu is similar to the Visualization menu, but is designed to verify the results after the imputation of missing values.
Last, but not least, the Options menu allows to set the colors, alpha
channel and the delimiter for imputed variables to be used in the plots. In
addition, it contains an option to embed multivariate plots in
windows. This is useful if the number of observations and/or variables is
large, because scrollbars allow to move from one part of the plot to
Internal information regarding the VIM GUI is stored in the environment
Andreas Alfons, based on an initial design by Matthias Templ, modifications by Bernd Prantner
M. Templ, A. Alfons, P. Filzmoser (2012) Exploring incomplete data using visualization tools. Journal of Advances in Data Analysis and Classification, Online first. DOI: 10.1007/s11634-011-0102-y.
A. Kowarik, M. Templ (2016) Imputation with R package VIM. Journal of Statistical Software, 74(7), 1-16
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.