Data-Driven Default Parameters for Regularized Normal Mixtures

Determines the default values for regularized univariate normal mixtures as proposed by Fraley and Raftery (2007) based on the data set to be clustered and the number of components in the mixture model.

FLXMCregnorm_defaults(x, zeta_p = NULL, kappa_p = 0.01, nu_p = 3, k = NULL)

Arguments

x

The data set to be clustered. Should be the same data set as is used in flexmix::flexmix()'s model formula.

zeta_p

Scale (hyperparameter for IG prior). If not given the empirical variance divided by the square of the number of components is used as per Fraley and Raftery (2007).

mu_p is computed from the data as the overall means across all components.

A value for the scale hyperparameter zeta_p may be specified directly. Otherwise the empirical variance divided by the square of the number of components is used as per Fraley and Raftery (2007). In which case the number of components (parameter k) needs to be specified.

kappa_p

Shrinkage parameter. Corresponds to adding kappa_p observations according to the population mean to each component (hyperparameter for IG prior)

nu_p

Degress of freedom (hyperparameter for IG prior)

k

Number of components assumed for the mixture model (not used if zeta_p is given).

Value

A named list with values for mu_p, kappa_p, nu_p and zeta_p.