EBSPriorProfiles: Matrix of prior values for each interval

In EBS: Exact Bayesian Segmentation

Description

Computes the matrix of indexes nr (values associated to prior on segmentation) for each segment r.

Usage

EBSPriorProfiles(n=numeric(), K = 3, unif=TRUE)

Arguments

n	The size of the series.
K	A vector of size the number of series which elements are the maximum number of segments that will be considered for each profile.
unif	A boolean stating whether the uniform prior will be used for the segmentation.

Details

This function is used to compute the values associated with the prior on the segmentation to use in computations such as ICL.

Value

An object of class EBSProfiles with values associated to prior on segmentation.

Author(s)

Alice Cleynen

References

Rigaill, Lebarbier & Robin (2012): Exact posterior distributions over the segmentation space and model selection for multiple change-point detection problems Statistics and Computing

Cleynen & Robin (2014): Comparing change-point location in independent series Statistics and Computing

See Also

EBSProfiles, EBSPrior

Examples

outPrior <- EBSPriorProfiles(1000,K=c(4,5,5),unif=FALSE)

EBS documentation built on May 29, 2017, 5:49 p.m.