mnemh {mnem} | R Documentation |
This function does a hierarchical mixture. That means it uses the approximate BIC to check, if there are more than one component. It recursively splits the data if there is evidence for k > 1 components.
mnemh(data, k = 2, logtype = 2, getprobspars = list(), ...)
data |
data matrix either binary or log odds |
k |
number of maximal components for each hierarchy leaf |
logtype |
log type of the data |
getprobspars |
list of parameters for the getProbs function |
... |
additional parameters for the mnem function |
object of class mnem
Martin Pirkl
sim <- simData(Sgenes = 3, Egenes = 2, Nems = 2, mw = c(0.4,0.6)) data <- (sim$data - 0.5)/0.5 data <- data + rnorm(length(data), 0, 1) result <- mnemh(data, starts = 1, k = 1)