The R package vismi (Visual Diagnostics for Multiple
Imputation) provides a
comprehensive suite of visual diagnostics for assessing the quality of
multiply imputed data. The package supports imputed data generated by
various multiple imputation methods, including mixgb, mice, and
more.
The package has been completely redesigned. Please check Articles section for more details.
-
vismi(): visualise multiply-imputed missing data through distributional characteristics -
vismi_overimp(): visualise multiply-imputed missing data through overimputation -
vismi_converge(): visualise convergence diagnostics for imputed values of an incomplete variable
With the support of trelliscopejs, vismi provides trelliscope displays to inspect all variables at once via
-
trellis_vismi() -
trellis_vismi_overimp() -
trellis_vismi_converge()
You can install the current development version of vismi from GitHub with:
devtools::install_github("agnesdeng/vismi")