vmGUImenu()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
Tcl/Tk 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
another.
Internal information regarding the VIM GUI is stored in
the environment vmGUIenvir.