disclapmix_robust: Robust fitting

In mikldk/disclapmix: Discrete Laplace Mixture Inference using the EM Algorithm

Robust fitting

Description

A wrapper around 'disclapmix()' that tries to avoid errors. Can sometimes avoid errors with SVD problems happening with ‘glm_method = ’internal_coef'‘ and 'glm_method = ’internal_dev''. This is done by taking a random subset of observations.

Usage

disclapmix_robust(
  x,
  clusters,
  rnd_obs = min(5L * clusters, round(nrow(x)/2)),
  rnd_tries = 10,
  ...
)

Arguments

x	Dataset.
clusters	The number of clusters/components to fit the model for.
rnd_obs	Number of random observations in subset if initial fit fails
rnd_tries	Number of tries with random subset if initial fit fails
...	Passed on to 'disclapmix()'

Examples

data(danes)
db <- as.matrix(danes[rep(1:nrow(danes), danes$n), 1:(ncol(danes)-1)])
fit <- disclapmix_robust(db, 3L)
fit

mikldk/disclapmix documentation built on Aug. 22, 2023, 10:56 a.m.