lstseq.cart: Calculates a scale of CART histograms

In delt: Estimation of Multivariate Densities Using Adaptive Partitions

Description

Calculates a scale of CART histograms. The histograms in the scale have partitions of growing cardinality. Returns a sequence of estimates as piecewise constant functions and optionally as level set trees. Optionally a shape tree for a level set of each estimate is calculated.

Usage

lstseq.cart(treeseq, maxleaf=NULL, lstree=NULL, level = NULL,
indvec = NULL)

Arguments

treeseq	output of function "prune"
maxleaf	the maximal cardinality of the partitions of the histograms in the sequence
lstree	if NULL, then level set trees are not calculated
level	if NULL, then shape trees are not calculated; if positive number, then it is the level of the level sets for which the shape trees are calculated
indvec	a vector of indeces; chooses a subset of the complete sequence of subhistograms of the overfitting histogram

Value

A list with components

lstseq	a list of level set trees
pcfseq	a list of piecewise constant functions
stseq	a list of shape trees
hseq	a vector of smoothing parameters corresponding to the members in the sequences; the smoothing parameter is the cardinality of the partition

Author(s)

Jussi Klemela

See Also

densplit, prune

Examples

library(denpro)
dendat<-sim.data(n=100,seed=1,type="mulmodII")
et<-densplit(dendat)
treeseq<-prune(et)

estiseq<-lstseq.cart(treeseq,maxleaf=20,lstree=TRUE)

mt<-modegraph(estiseq)

plotmodet(mt)

#scaletable(estiseq)

delt documentation built on May 2, 2019, 3:42 p.m.