initialize_ecm: Starting values for ECM

In ben-j-barlow/rcbmm: Regularized Copula-Based Mixture Model: Estimation and Model Selection

Description

Identifies optimal starting values for ECM using the results of MBHAC applied to the data. The data undergoes transformations to enhance separation amongst groups prior to performing MBHAC.

Usage

initialize_ecm(
  x,
  K,
  margins,
  transform = FALSE,
  hc_pairs = NULL,
  classification = NULL,
  trace = TRUE
)

Arguments

x	A numeric matrix or data frame of observations. Rows correspond to observations and columns correspond to variables.
K	The number of mixture componets.
margins	A character vector specifying the marginal distributions of the components in the mixture. The vector must have a length equal to the number of columns in x. Each element must be equal to "norm", "beta" or "gamma".
transform	A logical value indicating whether or not starting values should be obtained using the transformations SPH, PCS, PCR and SVD. The default is TRUE.
hc_pairs	The results from MBHAC obtained from the function call hc() using the mclust package. If NULL, the function obtains the results by calling hc().
classification	A numeric vector representing a partitioning of the data x. If not NULL, the clustering identified by the vector takeß preference over the clustering identified by hc_pairs.
trace	A logical value indicating if an update regarding the initialization procedure's progress should be displayed.

Value

• mixing_probsA numeric vector indicating the starting values of the mixing proportions.

• mvdcA list of objects of class mvdc. Each element of the list corresponds to a mixture component and contains the starting values for the component distribution's marginal and copula parameters.

• transformationA character string indicating the transformation applied prior to performing MBHAC.

• loglikThe log-likelihood corresponding of the model given the starting values of the parameters and the data.

See Also

hc

ben-j-barlow/rcbmm documentation built on Feb. 12, 2021, 9:14 a.m.