Deprecated lifecycle

Additional options for computing penalized EN S-estimates. Superseded by mm_algorithm_options() and options supplied directly to pense().

pense_options(
  delta = 0.25,
  maxit = 1000,
  eps = 1e-06,
  mscale_eps = 1e-08,
  mscale_maxit = 200,
  verbosity = 0,
  cc = NULL,
  en_correction = TRUE
)

Arguments

delta

desired breakdown point of the resulting estimator.

maxit

maximum number of iterations allowed.

eps

numeric tolerance for convergence.

mscale_eps, mscale_maxit

maximum number of iterations and numeric tolerance for the M-scale.

verbosity

ignored. Verbosity of the algorithm.

cc

ignored. Tuning constant for the S-estimator. Default is to chosen based on the breakdown point delta. Should never have to be changed.

en_correction

ignored. Should the corrected EN estimator be used to choose the optimal lambda with CV. If TRUE, as by default, the estimator is "bias corrected".

Warning

Do not use this function in new code. It may be removed from future versions of the package.

See also

Other deprecated functions: deprecated_en_options, enpy(), initest_options(), mstep_options(), pensem()