Tuning parameters for multivariate S, MM and GS estimates
Tuning parameters for multivariate S, MM and GS estimates
Tuning parameters for multivariate S, MM and GS estimates as used in FRB functions for multivariate regression, PCA and Hotelling tests. Mainly regarding the fast-(G)S algorithm.
nsamp: number of random subsamples to be used in the fast-(G)S algorithm
k: number of initial concentration steps performed on each subsample candidate
bestr: number of best candidates to keep for full iteration (i.e. concentration steps until convergence)
convTol: relative convergence tolerance for estimates used in (G)S-concentration iteration
maxIt: maximal number of steps in (G)S-concentration iteration
bdp: breakdown point of the MM-estimates; usually equals 0.5
eff: Gaussian efficiency of the MM-estimates; usually set at 0.95
shapeEff: logical; if TRUE, eff is with regard to shape-efficiency, otherwise location-efficiency
convTol.MM: relative convergence tolerance for estimates used in MM-iteration
maxIt.MM: maximal number of steps in MM-iteration
fastScontrols: the tuning parameters of the initial S-estimate
...: allows for any individual parameter from Scontrol to be set directly
Details
The default number of random samples is lower for GS-estimates than for S-estimates, because computations regarding the former are more demanding.
Returns
A list with the tuning parameters as set by the arguments.
References
S. Van Aelst and G. Willems (2013), Fast and robust bootstrap for multivariate inference: The R package FRB. Journal of Statistical Software, 53 (3), 1--32. tools:::Rd_expr_doi("10.18637/jss.v053.i03") .