kmeans.ct {ctmva}

Continuous-time k-means clustering

Description

A continuous-time version of k-means clustering in which each cluster is a time segments or set of time segments.

Usage

kmeans.ct(
  fdobj,
  k,
  common_trend = FALSE,
  init.pts = NULL,
  tol = 0.001,
  max.iter = 100
)

Arguments

fdobj

continuous-time multivariate data set of class "fd"

k

number of clusters

common_trend

logical: Should the curves be centered with respect to the mean function? Defaults to FALSE.

init.pts

a set of k time points. The observations at these time points serve as initial values for the k means. Randomly generated if not supplied.

tol

convergence tolerance for the k means

max.iter

maximum number of iterations

Value

Object of class "kmeans.ct", a list consisting of

fdobj

the supplied fdobj

means

means of the k clusters

transitions

transition points between segments

cluster

cluster memberships in the segments defined by the transitions

size

length of each cluster, i.e. sum of lengths of subintervals making up each cluster

totisd

total integrated sum of distances from the overall mean; this is the analogue of totss from link{kmeans}

withinisd

within-cluster integrated sum of distances, i.e. integrated sum of distances from each cluster mean

tot.withinisd

total within-cluster integrated sum of distances, i.e. sum(withinisd)

betweenisd

between-cluster integrated sum of distances, i.e. totisd-tot.withinss

Author(s)

Biplab Paul <[e-mail]> and Philip Tzvi Reiss <[e-mail]>

See Also

kmeans, plot.kmeans.ct

Examples

## Not run: 

require(fda)
data(CanadianWeather)
daybasis <- create.bspline.basis(c(0,365), nbasis=55)
tempfd <- smooth.basis(day.5, CanadianWeather$dailyAv[,,"Temperature.C"], daybasis)$fd
kmtemp3 <- kmeans.ct(tempfd, 3)
plot(kmtemp3, axes=FALSE)
axesIntervals(); box()
plot(silhouette.ct(kmtemp3), axes=FALSE)
axesIntervals(); box()

## End(Not run)