costMatrix {tilingArray}R Documentation

Segmentation cost matrix

Description

This function calculates the cost matrix for the segmentation model

Usage

costMatrix(x, maxk)

Arguments

x Numeric vector of length n or matrix with n rows and d columns, where n is the number of sample points and d the number of replicate measurements (e.g. from multiple arrays).
maxk Positive integer.

Details

See the vignette costMatrix.tex (.pdf) in the inst/doc directory of the package.

Value

Matrix with maxk rows and length(x) columns.

Author(s)

W. Huber

Examples

d = 4
x = apply(matrix(rnorm(200), ncol=d), 2, cumsum)
maxk = 50
 
G = costMatrix(x, maxk=maxk)
 
G.pedestrian = matrix(NA, nrow=nrow(G), ncol=ncol(G))
for(i in 1:(ncol(G)))
  for(k in 1:min(nrow(G), nrow(x)-i+1))
    G.pedestrian[k, i] = (k*d-1)*var(as.vector(x[i:(i+k-1), ]))
 
stopifnot(identical(is.na(G), is.na(G.pedestrian)))
stopifnot(max(abs(G-G.pedestrian), na.rm=TRUE) <= 1e-6)

[Package tilingArray version 1.10.0 Index]