Skip to contents

Distance-Distance plot for interval-valued data.

Source: R/plots.r
plot_dist_dist.Rd

Distance-Distance plot for interval-valued data.

Usage

plot_dist_dist(
  class_dist,
  class_cutoff = NULL,
  class_cutoff_label = NULL,
  rob_dist,
  rob_cutoff = NULL,
  rob_cutoff_label = NULL,
  obs_names = NULL,
  ggplotly = TRUE,
  color_class = NULL,
  color_label = NULL,
  palette = NULL,
  shape_class = NULL,
  shape_label = NULL,
  label_obs = NULL
)

Arguments

class_dist

A numeric vector containing the classical distances for each observation.

class_cutoff

Numeric. The cutoff value for the classical distances.

class_cutoff_label

Character. Label for the classical cutoff. If NULL (default), no legend for the classical cutoff is shown.

rob_dist

A numeric vector containing the robust distances for each observation.

rob_cutoff

Numeric. The cutoff value for the robust distances.

rob_cutoff_label

Character. Label for the robust cutoff. If NULL (default), no legend for the robust cutoff is shown.

obs_names

A character vector containing the names of the observations. If NULL (default), the names are taken from the names of class_dist.

ggplotly

Logical. If TRUE (default), the plot is converted to an interactive plotly::plotly object.

color_class

A vector indicating the color class of each observation. If NULL (default), all points have the same color.

color_label

Character. Label for the color class. If NULL (default), no legend for the color class is shown.

palette

A vector with colors for each color class. If NULL (default), default ggplot2::ggplot2 colors are used.

shape_class

A vector indicating the shape class of each observation. If NULL (default), all points have the same shape.

shape_label

Character. Label for the shape class. If NULL (default), no legend for the shape class is shown.

label_obs

A vector with the names of the observations to be labeled in the plot when ggplotly = FALSE. Default is NULL.

Value

Returns a Distance-Distance plot that displays the classical distances against the robust distances for each observation, highlighting outliers.

Examples

#Create intData object
data(creditcard)
credit_card_int <- creditcard$intData

#Estimate the mean and covariance matrix
credit_card_IMCD<-IMCD(credit_card_int, floor(nrow(credit_card_int)*0.75), "farness", 0.9)
credit_card_outliers <- int_outliers(credit_card_IMCD$robust_dist, 
                                           p=credit_card_int@NIVar, cutoff_lvl = 0.9)

#Plot Distance-Distance plot
class_dist <- IMah_dist(credit_card_int, z=rep(1,credit_card_int@NObs))
class_outliers <- int_outliers(class_dist,cutoff = "adjbox",p=p,cutoff_lvl = 1.5)
credit_card_is_outliers <- as.character(credit_card_outliers$is_outlier)
credit_card_is_outliers[credit_card_outliers$is_outlier] <- "Outlier"
credit_card_is_outliers[!credit_card_outliers$is_outlier] <- "Inlier"
plot_dist_dist(class_dist, class_outliers$cutoff_value[2], "1.5 adjusted boxplot",
              credit_card_IMCD$robust_dist, credit_card_outliers$cutoff_value, "0.9 farness",
              color_class = credit_card_is_outliers, palette = c("grey50", "red"))