DPEM {segclust}R Documentation

Dynamic Programing / EM segmentation/clustering

Description

Estimation of segmentation/clustering parameters in the Gaussian case, using dynamic programming and the EM algorithm. This function performs a global analysis with estimation and model selection. It uses functions hybrid(), segclustselect(), and seclustout(). Pmin and Pmax must be different.

Usage

        output <- DPEM(x,Pmin,Pmax,Kmax,method,draw,lmin=1,lmax=length(x),vh=TRUE,S=0.5)

Arguments

x data vector (without missing values)
Pmin minimum number of clusters
Pmax maximum number of clusters
Kmax max number of segments. Kmax must be greater than Pmax
method model selection method. Equals "sequential" or "BIC"
draw equals TRUE for a graphical display
lmin minimum segment length, default value lmin = 1
lmax maximum segment length, default value lmax = length(x)
vh TRUE for homogeneous variances (default), FALSE otherwise
S Threshold for model selection, set at 0.5

Value

output dataframe containing results of the estimation procedure
output$signal input signal x
output$mean estimated mean according to the model, for each position
output$sd estimated standard deviation according to the model, for each position
output$cluster cluster for each point
output$bp breakpoint coordinates, equals 1 for a breakpoint (corresponding to the end of the segments)

Author(s)

F. Picard, M. Hoebecke

References

Picard, F., Robin, S., Lebarbier, E., & Daudin, J. -J. (2007). A segmentation/clustering model for the analysis of array CGH data. Biometrics, 63(3) 758-766

Examples

        x1         <- rnorm(20,0,1)
        x2         <- rnorm(30,2,1)
        x3         <- rnorm(10,0,1)
        x4         <- rnorm(40,2,1)
        x          <- c(x1,x2,x3,x4)
        Pmin       <- 1
        Pmax       <- 4
        Kmax       <- 20
        output     <- DPEM(x,Pmin,Pmax,Kmax,method="BIC",draw=TRUE)
        

[Package segclust version 0.74 Index]