stab.blockSeg: 'stab.blockSeg' algorithm

In blockseg: Two Dimensional Change-Points Detection

Description

Model selection for the blockSeg algorithm.

Usage

stab.blockSeg(Y, nsimu, max.break, max.var = floor(ncol(Y)^2/8),
  random.break = TRUE, sym.break = FALSE, mc.cores = 2,
  verbose = TRUE)

Arguments

Y	matrix of observations.
nsimu	a positive integer.
max.break	a positive integer less than number of columns divided by 2 and number of rows divided by 2.
max.var	a positive integer less than number of columns times number of rows. By default, ncol(Y)**2/8.
random.break	logical. To change the position of the first row (resp. column); the rows before this position are moved to the end. By default TRUE.
sym.break	logical. In the case of symmetric matrices, it is possible to accumulate breaks in row and columns to improve the quality of the estimation. By default FALSE. Warning: a check is made on the dimensions of the matrix but not on the fact that it is symmetrical or not; this choice was made for the case where the user would like to have symmetrical breaks even if the matrix is not (not recommended by the authors of the package).
mc.cores	a positive integer giving the number of cores used. If you use windows, the parallelization is impossible. By default, 2
verbose	logical. To display each step. By default TRUE.

Examples

 ## model parameters
n <- 100
K <- 5
mu <- suppressWarnings(matrix(rep(c(1,0),ceiling(K**2/2)), K,K))
Y <- rblockdata(n,mu,sigma=.5)$Y
res <- stab.blockSeg(Y, 100, 20)

blockseg documentation built on May 2, 2019, 6:10 a.m.