getInfClipRegTS: Generic Function for the Computation of the Optimal Clipping...

In ROptRegTS: Optimally Robust Estimation for Regression-Type Models

Description

Generic function for the computation of the optimal clipping bound/function. This function is rarely called directly. It is used to compute optimally robust ICs in case infinitesimal models.

Usage

getInfClipRegTS(clip, ErrorL2deriv, Regressor, risk, neighbor, ...)

## S4 method for signature 
## 'numeric,UnivariateDistribution,Distribution,asMSE,Neighborhood'
getInfClipRegTS(
             clip, ErrorL2deriv, Regressor, risk, neighbor, z.comp, stand, cent)

## S4 method for signature 
## 'numeric,
##   UnivariateDistribution,
##   Distribution,
##   asMSE,
##   Av1CondTotalVarNeighborhood'
getInfClipRegTS(
             clip, ErrorL2deriv, Regressor, risk, neighbor, z.comp, stand, cent)

## S4 method for signature 
## 'numeric,EuclRandVariable,Distribution,asMSE,Neighborhood'
getInfClipRegTS(
             clip, ErrorL2deriv, Regressor, risk, neighbor, ErrorDistr, stand,
             cent, trafo)

## S4 method for signature 
## 'numeric,
##   UnivariateDistribution,
##   UnivariateDistribution,
##   asUnOvShoot,
##   UncondNeighborhood'
getInfClipRegTS(
             clip, ErrorL2deriv, Regressor, risk, neighbor, z.comp, cent)

## S4 method for signature 
## 'numeric,UnivariateDistribution,numeric,asUnOvShoot,CondNeighborhood'
getInfClipRegTS(
             clip, ErrorL2deriv, Regressor, risk, neighbor)

Arguments

clip	optimal clipping bound.
ErrorL2deriv	L2-derivative of ErrorDistr.
Regressor	regressor.
risk	object of class "RiskType".
neighbor	object of class "Neighborhood".
...	additional parameters.
cent	optimal centering constant/function.
stand	standardizing matrix.
z.comp	which components of the centering constant/function have to be computed.
ErrorDistr	error distribution.
trafo	matrix: transformation of the parameter.

Value

The optimal clipping bound/function is computed.

Methods

clip = "numeric", ErrorL2deriv = "UnivariateDistribution", Regressor = "Distribution", risk = "asMSE", neighbor = "Neighborhood"

optimal clipping bound for asymtotic mean square error.

clip = "numeric", ErrorL2deriv = "UnivariateDistribution", Regressor = "Distribution", risk = "asMSE", neighbor = "Av1CondTotalVarNeighborhood"

optimal clipping bound for asymtotic mean square error.

clip = "numeric", ErrorL2deriv = "EuclRandVariable", Regressor = "Distribution", risk = "asMSE", neighbor = "Neighborhood"

optimal clipping bound for asymtotic mean square error.

clip = "numeric", ErrorL2deriv = "UnivariateDistribution", Regressor = "UnivariateDistribution", risk = "asUnOvShoot", neighbor = "UncondNeighborhood"

optimal clipping bound for asymtotic under-/overshoot risk.

clip = "numeric", ErrorL2deriv = "UnivariateDistribution", Regressor = "numeric", risk = "asUnOvShoot", neighbor = "CondNeighborhood"

optimal clipping function for asymtotic under-/overshoot risk.

Author(s)

Matthias Kohl [e-mail]

References

Rieder, H. (1980) Estimates derived from robust tests. Ann. Stats. 8: 106–115.

Rieder, H. (1994) Robust Asymptotic Statistics. New York: Springer.

Kohl, M. (2005) Numerical Contributions to the Asymptotic Theory of Robustness. Bayreuth: Dissertation.

See Also

ContIC-class, TotalVarIC-class, Av1CondContIC-class, Av2CondContIC-class, Av1CondTotalVarIC-class, CondContIC-class, CondTotalVarIC-class

ROptRegTS documentation built on May 2, 2019, 3:40 p.m.