Type: Package
Title: Multivariate Analogy of Quantiles
Version: 0.1
Date: 2015-12-19
Maintainer: Somedip Karmakar <somedip@yahoo.co.in>
Description: Could be used to obtain spatial depths, spatial ranks and outliers of multivariate random variables. Could also be used to visualize DD-plots (a multivariate generalization of QQ-plots).
Imports: mvtnorm , stats , graphics
License: GPL-2
RoxygenNote: 5.0.1
NeedsCompilation: no
Packaged: 2015-12-20 11:01:29 UTC; omahala
Author: Omker Mahalanobish [aut], Somedip Karmakar [cre, aut]
Repository: CRAN
Date/Publication: 2015-12-20 15:33:19

Depth-Depth Plots

Description

dd.plot is a multivariate genralization of a normal QQ-plot. It produces a DD-plot of two datasets.

Usage

dd.plot(data1, data2 = rmvnorm(nrow(data1), array(0, ncol(data1)), diag(1,
  ncol(data1), ncol(data1))), main = "Normal DD-plot",
  xlab = "Sample Depths", ylab = "Normal Depths", col = "black",
  pch = 20)

Arguments

data1

A matrix or a data.frame with each row as a p-variate observation.

data2

A matrix or a data.frame (defaults to a standard independent p-variate normal).

main

Plot labels. The title of the plot.

xlab

Plot labels. The x-axis label of the plot.

ylab

Plot labels. The y-axis label of the plot.

col

The color of the points

pch

character string or vector of 1-characters or integers for plotting characters.

Value

A DD-plot of the input data

Author(s)

Somedip Karmakar <somedip@yahoo.co.in>

Omker Mahalanobish <omker.scorpio@gmail.com>

See Also

spatial.depth

Examples

u<-matrix(rnorm(300,1,4),ncol=3)
dd.plot(u)

Multivariate Quantile

Description

Used to compute the p-variate quantile of a p-variate observation with respect to a p-variate data cloud.

Usage

multi.quant(x, data)

Arguments

x

A numeric p-variate spatial rank. Elements must lie within -1 and +1, with a 0-vector denoting the median.

data

A matrix or a data.frame with each row as a p-variate observation.

Value

The xth mutivariate quantile with respect to data.

Author(s)

Somedip Karmakar <somedip@yahoo.co.in>

Omker Mahalanobish <omker.scorpio@gmail.com>

See Also

spatial.rank

Examples

u<-matrix(rnorm(90,0,1),ncol=3)
u0<-runif(3,0,1)
multi.quant(spatial.rank(u0,u),u)

Spatial Depth

Description

spatial.depth is used to find the spatial depth of one or more p-variate observation(s) in a data cloud of numerous p-variate observations.

Usage

spatial.depth(x, data)

Arguments

x

A matrix or a data.frame of objects (numerical vector as one object) whose depth is to be found; each row consists a p-variate observation.

data

A matrix or a data.frame of objects which acts as the data cloud. Each row consists of a p-variate observation.

Value

Numerical vector of depths, one for each row in x; or one depth value if x is numerical.

Author(s)

Somedip Karmakar <somedip@yahoo.co.in>

Omker Mahalanobish <omker.scorpio@gmail.com>

Examples

u<-matrix(rnorm(90,0,1),ncol=3)
u0<-matrix(runif(9,0,1),ncol=3)
spatial.depth(u0,u)

Multivariate Spatial Outlier

Description

spatial.outlier is used to find the multivariate spatial outlier within a p-variate data cloud or to identify if any p-variate observation is an outlier with respect to a p-variate data cloud.

Usage

spatial.outlier(data, x = data, threshold = 0.05)

Arguments

data

A matrix or a data.frame of p-variate observations which works as the data cloud.

x

A matrix or a data.framep-variate to test whether is an outlier with respect to the data. Defaults to data, to find outliers (if exists) within the data.

threshold

A decimal threshold between 0 and 1 on the spatial.depth. Spatial depth values less than which will be considered as outlier. Defaults to 0.05. Usually taken as 0.1 or 0.05 or 0.01.

Value

FALSE :: If there doesnot exist any outlier

A list with objects (If outliers exist)

index :: Returns the indices of the outliers

observation :: Returns the p-variate outliers

Author(s)

Somedip Karmakar <somedip@yahoo.co.in>

Omker Mahalanobish <omker.scorpio@gmail.com>

Examples

u<-matrix(rnorm(60,0,1),ncol=3)
u0<-matrix(runif(9,3,4),ncol=3)
spatial.outlier(u,rbind(u,u0))

Spatial Rank

Description

Used to compute the Spatial Rank of a p-variate observation with respect to a p-variate data cloud.

Usage

spatial.rank(x, data)

Arguments

x

A numeric p-variate vector whose spatial rank is to be calculated.

data

A matrix or a data.frame with each row as a p-variate observation.

Value

The spatial rank of x with respect to data.

Author(s)

Somedip Karmakar <somedip@yahoo.co.in>

Omker Mahalanobish <omker.scorpio@gmail.com>

Examples

u<-matrix(rnorm(90,0,1),ncol=3)
u0<-runif(3,0,1)
spatial.rank(u0,u)