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Superseded by mm_algorithm_options()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 )
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 |
en_correction | ignored. Should the corrected EN estimator be used to choose
the optimal lambda with CV.
If |
Do not use this function in new code. It may be removed from future versions of the package.
Other deprecated functions:
deprecated_en_options
,
enpy()
,
initest_options()
,
mstep_options()
,
pensem()