ssic.penalty: Strengthened Schwarz Information Criterion penalty term

In wbs: Wild Binary Segmentation for Multiple Change-Point Detection

Description

The function evaluates the penalty term for the strengthened Schwarz Information Criterion proposed in P. Fryzlewicz (2014). This routine is typically not called directly by the user; its name can be passed as an argument to changepoints.

Usage

ssic.penalty(n, cpt, alpha = 1.01, ssic.type = c("log", "power"))

Arguments

n	the number of observations
cpt	a vector with localisations of change-points
alpha	a scalar greater than one
ssic.type	a string ("log" or "power")

Value

the penalty term k(log(n))^(alpha) for ssic.penalty="log" or k * n^(alpha) for ssic.penalty="power", where k denotes the number of elements in cpt

References

P. Fryzlewicz (2014), Wild Binary Segmentation for multiple change-point detection. Annals of Statistics, to appear. (http://stats.lse.ac.uk/fryzlewicz/wbs/wbs.pdf)

Examples

x <- rnorm(300) + c(rep(1,50),rep(0,250))
w <- wbs(x)
w.cpt <- changepoints(w,penalty="ssic.penalty")
w.cpt$cpt.ic

Example output

$ssic.penalty
[1] 54

wbs documentation built on May 15, 2019, 1:04 a.m.