| sic.penalty {not} | R Documentation |
Schwarz Information Criterion penalty
Description
The function evaluates the penalty term for Schwarz Information Criterion.
If alpha is greater than 1, the strengthen SIC proposed proposed in Fryzlewicz (2014) is calculated. This routine is typically not called directly by the user;
its name can be passed as an argument to features.
Usage
sic.penalty(n, n.param, alpha = 1, ...)
Arguments
n |
The number of observations. |
n.param |
The number of parameters in the model for which the penalty is evaluated. |
alpha |
A scalar greater or equal than one. |
... |
Not in use. |
Value
the penalty term \code{n.param}\times(\log(n))^{\code{alpha}}.
References
R. Baranowski, Y. Chen, and P. Fryzlewicz (2019). Narrowest-Over-Threshold Change-Point Detection. (http://stats.lse.ac.uk/fryzlewicz/not/not.pdf)
P. Fryzlewicz (2014). Wild Binary Segmentation for multiple change-point detection. Annals of Statistics. (http://stats.lse.ac.uk/fryzlewicz/wbs/wbs.pdf)
Examples
#*** a simple example how to use the AIC penalty
x <- rnorm(300) + c(rep(1,50),rep(0,250))
w <- not(x)
w.cpt <- features(w, penalty="sic")
w.cpt$cpt[[1]]