EBSICLProfiles: Model Selection by Integrated Completed Likelihood criterion
In EBS: Exact Bayesian Segmentation

Description Usage Arguments Details Value Author(s) References See Also Examples

For each profile, computes the exact ICL criterion: -Loglikelihood (data,K) + H(m|K) where H is the entropy of the segmentation, and chooses the optimal number of segments as k= argmin(ICL)

1	EBSICLProfiles(x, prior=numeric())

`x`	An object of class EBSPofiles returned by function EBSProfiles applied to matrix of profiles of interest.
`prior`	A vector of length Kmax giving prior probabilities on the value of K. Default value is uniform on 1:Kmax.

For each condition, this function is used to compute the entropy of the segmentation in k segments (for k in 1 to Kmax) and choose the optimal K according to the ICL criteria.

`NbICL`	A vector containing the choice of the optimal number of segments for each profile.
`ICL`	A list of vector (one for each condition) of length getK(x)[l] containing the ICL values.

Alice Cleynen

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

EBSProfiles, EBSICL

# changes for Poisson model
set.seed(1)
x1<-c(rpois(125,1),rpois(100,5),rpois(50,1),rpois(75,5),rpois(50,1))
x2<-c(rpois(100,1),rpois(100,3),rpois(75,2),rpois(125,0.5))
M<-rbind(x1,x2)
E<-EBSProfiles(M, K=8)
out <- EBSICLProfiles(E)
print(out$NbICL)