EBS (version 3.1)

CompCredibility: Comparison of two profiles with credibility intervals

Description

Given two conditions, computes the posterior distribution of the difference of change-point locations, and its credibility interval.

Usage

CompCredibility(x, Conditions, Tau = numeric(), K = numeric())

Arguments

x

An object of class EBSProfiles.

Conditions

A vector of length 2 containing the index of the two conditions to compare.

Tau

A vector of length 2 containing the index of the change-point of interest of the two conditions to compare.

K

A vector of length 2 containing the maximum number of segments for the segmentation of each of the two profiles to compare.

Value

Distribution

Posterior distribution of the difference between the location of change-points of interest for the two conditions

masswith0

Mass of the smallest credibility interval up to and including data-point 0

massto0

Mass of the smallest credibility interval up to but excluding data-point 0

Details

This function is used to compute the posterior credibility interval of the difference of change-point locations between two profiles assumed to be independant.

References

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

Johnson, Kotz & Kemp: Univariate Discrete Distributions

Hall, Kay & Titterington: Asymptotically optimal difference-based estimation of variance in non-parametric regression

See Also

EBSProfiles, EBSStatistic

Examples

# 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(125,3),rpois(75,4),rpois(75,1),rpois(125,8))
M<-rbind(x1,x2)
E <- EBSProfiles(M,model=1,K=10)
out<-CompCredibility(E,Conditions=c(1,2),Tau=c(1,1),K=c(5,4))