Generates an object of class "CondTotalVarIC"
;
i.e., an influence curves \(\eta\) of the form
$$\eta = max(c(x), min(A x\Lambda_f, b(x)))$$
with lower clipping function \(c\), upper clipping function \(b\) and
standardizing matrix \(A\). \(\Lambda_f\) stands for
the L2 derivative of the corresponding error distribution.
CondTotalVarIC(name, CallL2Fam = call("L2RegTypeFamily"),
Curve = EuclRandVarList(RealRandVariable(
Map = list(function(x) {x[1] * x[2]}),
Domain = EuclideanSpace(dimension = 2))),
Risks, Infos,
clipUp = RealRandVariable(Map = list(function(x) {Inf}), Domain = Reals()),
stand = as.matrix(1),
clipLo = RealRandVariable(Map = list(function(x) {-Inf}), Domain = Reals()),
lowerCase = NULL, neighborRadius = 0, neighborRadiusCurve = function(x){1})
object of class "character"
.
object of class "call"
:
creates an object of the underlying L2-differentiable
regression type family.
object of class "EuclRandVarList"
object of class "list"
:
list of risks; cf. RiskType-class
.
matrix of characters with two columns
named method
and message
: additional informations.
object of class "RealRandVariable"
:
upper clipping function.
object of class "RealRandVariable"
:
lower clipping function.
matrix: standardizing matrix.
optional constant for lower case solution.
radius of the corresponding conditional total variation neighborhood.
radius curve of the corresponding conditional total variation neighborhood.
Object of class "CondTotalVarIC"
Rieder, H. (1994) Robust Asymptotic Statistics. New York: Springer.
Kohl, M. (2005) Numerical Contributions to the Asymptotic Theory of Robustness. Bayreuth: Dissertation.
# NOT RUN {
IC1 <- CondTotalVarIC()
IC1
# }
Run the code above in your browser using DataLab