changepoints (version 1.1.0)

tuneBSuninonpar: Wild binary segmentation for univariate nonparametric change points detection with tuning parameter selection.

Description

Perform wild binary segmentation with tuning parameter selection based on sample splitting.

Usage

tuneBSuninonpar(BS_object, Y, N)

Value

A vector of estimated change points (sorted in strictly increasing order).

Arguments

BS_object

A "BS" object produced by BS.uni.nonpar or WBS.uni.nonpar.

Y

A numeric matrix of observations with horizontal axis being time, and vertical axis being multiple observations on each time point.

N

A integer vector representing number of multiple observations on each time point.

Author

Oscar Hernan Madrid Padilla & Haotian Xu

References

Padilla, Yu, Wang and Rinaldo (2021) <doi:10.1214/21-EJS1809>.

See Also

BS.uni.nonpar and WBS.uni.nonpar.

Examples

Y = t(as.matrix(c(rnorm(100, 0, 1), rnorm(100, 0, 10), rnorm(50, 0, 40))))
W = Y # W is a copy of the matrix Y, it can be Y itself.
N = rep(1, 250)
M = 5
set.seed(123)
intervals = WBS.intervals(M = M, lower = 1, upper = ncol(Y))
BS_object = WBS.uni.nonpar(W, 1, ncol(Y), intervals$Alpha, intervals$Beta, N, delta = 5)
cpt_hat = tuneBSuninonpar(BS_object, Y, N)