step.lsbclust {lsbclust}

Model Search for lsbclust

Description

Fit lsbclust models for different numbers of clusters and/or different values of delta. The resulting output can be inspected through its plot method to facilitate model selection. Each component of the model is fitted separately.

Usage

step.lsbclust(data, margin = 3L, delta = c(1, 1, 1, 1), nclust,
  ndim = 2, fixed = c("none", "rows", "columns"), nstart = 20,
  starts = NULL, nstart.kmeans = 500, alpha = 0.5,
  parallel = FALSE, maxit = 100, verbose = -1, type = NULL, ...)

Arguments

data

A three-way array representing the data.

margin

An integer giving the single subscript of data over which the clustering will be applied.

delta

A four-element binary vector (logical or numeric) indicating which sum-to-zero constraints must be enforced.

nclust

Either a vector giving the number of clusters which will be applied to each element of the model, that is to (a subset of) the overall mean, row margins, column margins and interactions. If it is a list, arguments are matched by the names "overall", "rows" "columns" and "interactions". If the list does not have names, the components are extracted in the aforementioned order.

ndim

The required rank for the approximation of the interactions (a scalar).

fixed

One of "none", "rows" or "columns" indicating whether to fix neither sets of coordinates, or whether to fix the row or column coordinates across clusters respectively. If a vector is supplied, only the first element will be used (passed to int.lsbclust).

nstart

The number of random starts to use for the interaction clustering.

starts

A list containing starting configurations for the cluster membership vector. If not supplied, random initializations will be generated (passed to int.lsbclust).

nstart.kmeans

The number of random starts to use in kmeans.

alpha

Numeric value in [0, 1] which determines how the singular values are distributed between rows and columns (passed to int.lsbclust).

parallel

Logical indicating whether to parallelize over different starts or not (passed to int.lsbclust).

maxit

The maximum number of iterations allowed in the interaction clustering.

verbose

The number of iterations after which information on progress is provided (passed to int.lsbclust).

type

One of "rows", "columns" or "overall" (or a unique abbreviation of one of these) indicating whether clustering should be done on row margins, column margins or the overall means of the two-way slices respectively. If more than one opion are supplied, the algorithm is run for all (unique) options supplied (passed to orc.lsbclust). This is an optional argument.

...

Additional arguments passed to kmeans.

Examples

m <- step.lsbclust(data = dcars, margin = 3, delta = c(1, 0, 1, 0), nclust = 4:5, 
                     ndim = 2, fixed = "columns", nstart = 1, nstart.kmeans = 100, 
                     parallel = FALSE)
                     
## For a list of all deltas                     
delta <- expand.grid(replicate(4, c(0,1), simplify = FALSE))
delta <- with(delta, delta[!(Var1 == 0 & Var3 == 1), ])
delta <- with(delta, delta[!(Var2 == 0 & Var4 == 1),])
delta <- delta[-4,]
delta <- as.list(as.data.frame(t(delta)))
m2 <- step.lsbclust(data = dcars, margin = 3, delta = delta, nclust = 4:5, 
                     ndim = 2, fixed = "columns", nstart = 1, nstart.kmeans = 100, 
                     parallel = FALSE)