mbcbigp: Fit a mixture of multivariate Gaussians using sequential...
In markajoc/MBCbigP: Model-Based Clustering for High-Dimensional Data

Description Usage Arguments Value Examples

Fit a mixture of multivariate Gaussians

mbcbigp(x, groups = 2, batches = NULL, batchindex = NULL,
  batchsize = NULL, maxiter = 50, plot = FALSE, likelihood = TRUE,
  verbose = TRUE, abstol = 1e-06, method_sigma_AB = c("analytic",
  "numeric"), z = NULL, scorefun = bestseparated, updateA = FALSE)

`x`	Data frame or a matrix
`groups`	The number of groups/mixture components to fit.
`batches`	The batches to split the columns of `x` for processing. A single number gives the desired number of batches, whereas a vector of length `ncol(x)` is treated as an integer index of batch membership.
`batchindex`	A list of column names or integer indices for the batches of `x`.
`batchsize`	The desired number of columns in each batch.
`maxiter`	The maximum number of iterations for the E-M algorithm.
`plot`	Logical, should a visualisation be produced to show progress.
`likelihood`	Logical, should the likelihood be calculated and used as a stopping criterion.
`verbose`	Logical, should statements about progress be printed.
`abstol`	Numeric, tolerance for log-likelihood stopping rule. Defaults to 1e-6.
`method_sigma_AB`	One of `"analytic"` (default) or `"numeric"`, to choose the method of estimating the between-batch covariance for a cluster.
`z`	Initial cluster membership probabilities. Numeric matrix.
`scorefun`	A function to score the clustering of the data at each batch, defaults to `bestseparated`.
`updateA`	If `FALSE` (default), at each batch the parameters for the current batch are maximised conditionally on the parameters from the previous batch. If `TRUE`, this is followed by also maximising the parameters from the previous batch conditionally on the current batch.

A list containing the estimated parameters for the mixture distribution.

## Not run: 
data(banknote, package = "mclust")
mbcbigp(x = banknote[, -1], groups = 2, batches = 2)
mbcbigp(x = banknote[, -1], groups = 2, batches = 3)

## End(Not run)