kda.kde: Kernel density estimate for discriminant analysis for multivariate data

Description

Kernel density estimate for discriminant analysis for 1- to 6-dimensional data.

Usage

kda.kde(x, x.group, Hs, hs, prior.prob=NULL, gridsize, xmin, xmax,
    supp=3.7, eval.points=NULL, binned=FALSE, bgridsize, w,
    compute.cont=FALSE, approx.cont=TRUE)

Arguments

matrix of data values

x.group

vector of group labels

bandwidth matrix(ces)

scalar bandwidth(s)

prior.prob

vector of prior probabilities

gridsize

vector of number of grid points

xmin

vector of minimum values for grid

xmax

vector of maximum values for grid

supp

effective support for standard normal.

eval.points

points at which density estimate is evaluated

binned

flag for binned estimation. Default is FALSE.

bgridsize

vector of binning grid sizes

vector of weights (non-negative and sum is equal to sample size). Default is a vector of all ones.

compute.cont

flag for computing probability contour levels from 1% to 99%. Default is FALSE.

approx.cont

flag for computing approximate probability contour levels. Default is TRUE.

Value

A density estimate for discriminant analysis is an object of class kda.kde:
xdata points - same as input
x.groupgroup labels - same as input
eval.pointspoints that density estimate is evaluated at
estimatedensity estimate at eval.points
prior.probprior probabilities
Hbandwidth matrices (>1-d only) or
hbandwidths (1-d only)
contprobability contour levels
wweights

Details

If you have prior probabilities then set prior.prob to these. Otherwise the default is prior.prob=NULL i.e. use the sample proportions as estimates of the prior probabilities. For details of the other parameters, see kde.

Examples

Run this code

## trivariate example
## colour indicates species, transparency indicates density heights
library(MASS)
data(iris)
ir <- iris[,1:3]
ir.gr <- iris[,5] 
H <- Hkda(ir, ir.gr, bw="plugin", pilot="dscalar")
kda.fhat <- kda.kde(ir, ir.gr, Hs=H)
plot(kda.fhat)