Inspection tab

The inspection tab offers users the options to create quantile-quantile-plots and visualize variable correlation pattern.

Quantile-Quantile plots

inspection_qq_plot

The purpose of quantile-quantile plots (QQ-plots) is to check whether a variable shows a normal distribution. If a variable is perfectly normally distributed each data point will follow the faint black diagonal which represents a theoretical normal distribution for the given data.

mvpaShiny allows to show several QQ-plots side-by-side, using the multiple selection field.

Correlation plots

inspection_correlation_single_response

mvpaShiny can plot two types of correlation:

  • Variables with (a single) response. The plot can be rotated by 90° for easier interpretation.
  • Variables with each other

inspection_correlation_matrix