ssic.penalty {wbs} | R Documentation |
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
.
ssic.penalty(n, cpt, alpha = 1.01, ssic.type = c("log", "power"))
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") |
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
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)
x <- rnorm(300) + c(rep(1,50),rep(0,250)) w <- wbs(x) w.cpt <- changepoints(w,penalty="ssic.penalty") w.cpt$cpt.ic