1\name{depth.space.} 2\alias{depth.space.} 3 4\title{ 5Calculate Depth Space using the Given Depth 6} 7\description{ 8Calculates the representation of the training classes in depth space. 9 10The detailed descriptions are found in the corresponding topics. 11} 12\usage{ 13depth.space.(data, cardinalities, notion, ...) 14 15## Mahalanobis depth 16# depth.space.Mahalanobis(data, cardinalities, mah.estimate = "moment", mah.parMcd = 0.75) 17 18## projection depth 19# depth.space.projection(data, cardinalities, method = "random", num.directions = 1000) 20 21## Tukey depth 22# depth.space.halfspace(data, cardinalities, exact, alg, num.directions = 1000) 23 24## spatial depth 25# depth.space.spatial(data, cardinalities) 26 27## zonoid depth 28# depth.space.zonoid(data, cardinalities) 29 30# Potential 31# depth.space.potential(data, cardinalities, pretransform = "NMom", 32# kernel = "GKernel", kernel.bandwidth = NULL, mah.parMcd = 0.75) 33} 34 35\arguments{ 36 \item{data}{ 37Matrix containing training sample where each row is a \eqn{d}-dimensional object, and objects of each class are kept together so that the matrix can be thought of as containing blocks of objects representing classes. 38} 39 \item{cardinalities}{ 40Numerical vector of cardinalities of each class in \code{data}, each entry corresponds to one class. 41} 42 \item{notion}{ 43The name of the depth notion (shall also work with \code{\link{Custom Methods}}). 44} 45 \item{\dots}{ 46Additional parameters passed to the depth functions. 47} 48} 49 50\value{ 51Matrix of objects, each object (row) is represented via its depths (columns) w.r.t. each of the classes of the training sample; order of the classes in columns corresponds to the one in the argument \code{cardinalities}. 52} 53 54\seealso{ 55 56\code{\link{depth.space.Mahalanobis}} 57 58\code{\link{depth.space.projection}} 59 60\code{\link{depth.space.halfspace}} 61 62\code{\link{depth.space.spatial}} 63 64\code{\link{depth.space.zonoid}} 65 66} 67\examples{ 68# Generate a bivariate normal location-shift classification task 69# containing 20 training objects 70class1 <- mvrnorm(10, c(0,0), 71 matrix(c(1,1,1,4), nrow = 2, ncol = 2, byrow = TRUE)) 72class2 <- mvrnorm(10, c(2,2), 73 matrix(c(1,1,1,4), nrow = 2, ncol = 2, byrow = TRUE)) 74data <- rbind(class1, class2) 75# Get depth space using zonoid depth 76depth.space.(data, c(10, 10), notion = "zonoid") 77} 78\keyword{ robust } 79\keyword{ multivariate } 80\keyword{ nonparametric } 81