Wrap Observations

• wrap_hull() computes convex hull of a set of observations.

• wrap_confidence() computes a confidence ellipse.

• wrap_tolerance() computes a tolerance ellipse.

Usage

wrap_hull(x, ...)

wrap_confidence(x, ...)

wrap_tolerance(x, ...)

# S4 method for MultivariateAnalysis
wrap_confidence(x, margin = 1, axes = c(1, 2), group = NULL, level = 0.95)

# S4 method for MultivariateAnalysis
wrap_tolerance(x, margin = 1, axes = c(1, 2), group = NULL, level = 0.95)

# S4 method for MultivariateAnalysis
wrap_hull(x, margin = 1, axes = c(1, 2), group = NULL)

Arguments

x

An object from which to wrap observations (a CA or PCA object).

...

Currently not used.

margin

A length-one numeric vector giving the subscript which the data will be returned: 1 indicates individuals/rows (the default), 2 indicates variables/columns.

axes

A length-two numeric vector giving the dimensions to be for which to compute results.

group

A vector specifying the group an observation belongs to.

level

A numeric vector specifying the confidence/tolerance level.

Value

wrap_*() returns a data.frame of envelope x and y coordinates.

An extra column named group is added specifying the group an observation belongs to.

See also

Other plot methods: biplot(), screeplot(), viz_contributions(), viz_individuals(), viz_variables(), viz_wrap

Author

N. Frerebeau

Examples

## Load data
data("iris")

## Compute principal components analysis
X <- pca(iris, scale = TRUE)
#> 1 qualitative variable was removed: Species.

## Convex hull coordinates
hulls <- wrap_hull(X, margin = 1, group = iris$Species)

## Confidence ellipse coordinates
conf <- wrap_confidence(X, margin = 1, group = iris$Species,
                        level = c(0.68, 0.95))

## Tolerance ellipse coordinates
conf <- wrap_confidence(X, margin = 1, group = iris$Species, level = 0.95)

## Plot with convex hulls
col <- c("#004488", "#DDAA33", "#BB5566")
viz_rows(X, highlight = iris$Species, col = col)
viz_hull(X, group = iris$Species, border = col)