NormLinRegScaleFamily: Generating Function for Linear Regression Family with Unknown...

In ROptRegTS: Optimally Robust Estimation for Regression-Type Models

Description

Generates an object of class "L2RegTypeFamily" which represents a linear regression family with standard normal distributed errors and random regressor where the scale of the error distribution is unknown.

Usage

NormLinRegScaleFamily(theta, scale = 1, trafo, RegDistr = Norm(), 
                      RegSymm, Reg2Mom, nuisance = FALSE)

Arguments

theta	linear regression parameter
scale	scale parameter for error distribution
trafo	matrix: transformation of the parameter
RegDistr	regressor distribution
RegSymm	symmetry of the regressor distribution
Reg2Mom	second moment matrix of regressor
nuisance	logical: is scale nuisance parameter

Details

In case theta is missing, it is set to 0. If Reg2Mom is missing, it is computed via E.

Value

Object of class "L2RegTypeFamily"

Author(s)

Matthias Kohl [e-mail]

References

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

See Also

L2RegTypeFamily-class

Examples

(LM1 <- NormLinRegScaleFamily(Reg2Mom = matrix(1)))
Map(L2deriv(LM1)[[1]])
FisherInfo(LM1)
checkL2deriv(LM1)

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