myStepFlexmix: Run 'flexmix' Repeatedly
In schiffner/locClass: Collection of Local Classification Methods

Description Usage Arguments Value See Also Examples

Runs flexmix repeatedly for different cluster values and returns the result with highest likelihood value.

1	myStepFlexmix(..., cluster, verbose = FALSE, drop = TRUE, unique = FALSE)

`cluster`	List of initial cluster assignments of observations at the start of the EM algorithm.
`verbose`	If `TRUE`, show progress information during computations.
`drop`	If `TRUE` and k is of length 1, then a single flexmix object is returned instead of a "stepFlexmix" object.
`unique`	If `TRUE`, then `unique()` is called on the result, see below.
`...`	Passed to `flexmix`.

An object of class "stepFlexmix" containing the best models with respect to the log likelihood for the different number of components in a slot if length(k)>1, else directly an object of class "flexmix". If unique=FALSE, then the resulting object contains one model per element of k (which is the number of clusters the EM algorithm started with). If unique=TRUE, then the result is resorted according to the number of clusters contained in the fitted models (which may be less than the number with which the EM algorithm started), and only the maximum likelihood solution for each number of fitted clusters is kept. This operation can also be done manually by calling unique() on objects of class "stepFlexmix".

Other mixtures: FLXMCL-class, FLXPwlda-class, FLXPwlda, myfitted, mypredict

library(benchData)
data <- xor3Data(500)
model <- FLXMCLmultinom(trace = FALSE, decay = 0.1)
cluster <- lapply(1:5, kmeans(data$x, centers = 3)$cluster)
fit <- myStepFlexmix(y ~ ., data = as.data.frame(data), concomitant = FLXPmultinom(~ x.1 + x.2), model = model, 
    cluster = cluster, control = list(verb = 1, tolerance = 10^-1))
fit2 <- flexmix(y ~ ., data = as.data.frame(data), concomitant = FLXPmultinom(~ x.1 + x.2), model = model, 
    cluster = posterior(fit), control = list(verb = 1))
pred <- mypredict(fit2, aggregate = TRUE)
mean(max.col(pred[[1]]) != data$y)