clustersegments {segclust} | R Documentation |
Cluster segments using the EM algorithm when the number of clusters and the segmentation are given. The segmentation is not revised after clustering (contrary to hybrid)
out <- clustersegments(x,P,bp,vh=TRUE)
x |
data vector (without missing values) |
P |
number of clusters |
bp |
vector (size length(x)), such that bp[t]=1 if t is a breakpoint and 0 otherwise. (t corresponds to the end of a segment). bp[length(x)] is always equal to 1. |
vh |
TRUE for homogeneous variances (default), FALSE otherwise |
out |
dataframe |
output$signal |
input signal x |
output$mean |
estimated mean using a mixture model with P cluster, AFTER a segmentation |
output$sd |
estimated standard deviation using a mixture model with P cluster, AFTER a segmentation |
output$cluster |
cluster for each point AFTER segmentation |
output$bp |
breakpoint coordinates, equals 1 for a breakpoint (corresponding to the end of the segments) |
F. Picard, M. Hoebecke
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
x1 <- rnorm(20,0,1) x2 <- rnorm(30,2,1) x <- c(x1,x2) bp <- rep(0,length(x)) bp[c(20,50)] <- 1 P <- 2 out <- clustersegments(x,P,bp)