EBSProfiles-class: Class "EBSProfiles"

In EBS: Exact Bayesian Segmentation

Description

A class for Bayesian Segmentation and comparison of multiple profiles.

Objects from the Class

Objects can be created by calls of the form new("EBSProfiles", ...).

new("EBSProfiles", ...):

creates a new object with class EBSProfiles

Slots

model:

Object of class "character", the assumed (identic) distribution class of each profile

data:

Object of class "numeric", the matrix of profiles (lines) to be segmented

length:

Object of class "numeric", the length of each profile

NbConditions:

Object of class "numeric", the number of profiles

K:

Object of class "numeric", the vector of maximum number of segments considered for each profile

HyperParameters:

Object of class "numeric", the vector of hyperparameters used for the prior distribution on the model parameters for each profile

Variance:

Object of class "numeric", if model = Normal Homoscedastic, the vector of variances used in the analysis

overdispersion:

Object of class "numeric", if model = Negative Binomial, the vector of overdispersions used in the analysis

Li:

Object of class "numeric", a list of the matrix of each profile, of size Kmax*(length+1) where element [i,j] is the log-probability of interval [1,j[ being segmented in j segments

Col:

Object of class "numeric", a list of the matrix of each profile, of size (length+1)*Kmax where element [i,j] is the log-probability of interval [i,n] being segmented in i segments

P:

a list of the matrix of each profile, of size (length+1)*(length+1) where element [i,j] is the log-probability of interval [i,j[

unif:

a boolean stating whether the prior on the segmentation is uniform given the number of segments.

Methods

Model

signature(object = "EBSProfiles"): retrieves model slot

Data

signature(object = "EBSProfiles"): retrieves data slot

Length

signature(object = "EBSProfiles"): retrieves length slot

NbConditions

signature(object = "EBSProfiles"): retrieves NbConditions slot

Kmax

signature(object = "EBSProfiles"): retrieves K slot

HyperParameters

signature(object = "EBSProfiles"): retrieves HyperParameters slot

Variance

signature(object = "EBSProfiles"): retrieves Variance slot

Overdispersion

signature(object = "EBSProfiles"): retrieves overdispersion slot

Li

signature(object = "EBSProfiles"): retrieves Li slot

Col

signature(object = "EBSProfiles"): retrieves Col slot

matProba

signature(object = "EBSProfiles"): retrieves P slot

Priorm

signature(object = "EBSProfiles"): retrieves unif slot

Author(s)

Alice Cleynen

See Also

EBSegmentation,EBSProfiles, Classes

Examples

showClass("EBS") # shows the structure of the cpt class

x1<-c(rpois(100,2),rpois(200,5))
x2<-c(rpois(100,3),rpois(150,8),rpois(50,2))
data<-rbind(x1,x2)
# creates a new EBSProfiles object containing the segmentation of 
#profiles x1 and x2
E<-EBSProfiles(data,K=c(2,3))
class(E) # verifies the class of E
Model(E) # retrieves model of the segmentation
# retrieves the maximal number of segments considered for profile x1
Kmax(E)[1]

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