OGKCStep {DetR}

Robust and Deterministic Linear Regression via OGKCStep

Description

Function to find the OGKCStep ('best') H-subset.

Usage

  OGKCStep(x0, scale_est, alpha=0.5)

Arguments

x0

Matrix of continuous variables.

alpha

numeric parameter controlling the size of the subsets over which the determinant is minimized, i.e., alpha*n observations are used for computing the determinant. Allowed values are between 0.5 and 1 and the default is 0.5.

scale_est

A character string specifying the variance functional. Possible values are Qn or scaleTau2.

Value

best

the best subset found and used for computing the raw estimates, with length(best) == quan = h.alpha.n(alpha,n,p).

Author(s)

Large part of the the code are from function .detmcd in package robustbase , , see citation("robustbase")

References

Maronna, R.A. and Zamar, R.H. (2002) Robust estimates of location and dispersion of high-dimensional datasets; Technometrics 44(4), 307–317.

Rousseeuw, P.J. and Croux, C. (1993) Alternatives to the Median Absolute Deviation; Journal of the American Statistical Association , 88(424), 1273–1283.

Peter J. Rousseeuw (1984), Least Median of Squares Regression. Journal of the American Statistical Association 79, 871–881.

P. J. Rousseeuw and A. M. Leroy (1987) Robust Regression and Outlier Detection. Wiley.

P. J. Rousseeuw and K. van Driessen (1999) A fast algorithm for the minimum covariance determinant estimator. Technometrics 41, 212–223.

Pison, G., Van Aelst, S., and Willems, G. (2002) Small Sample Corrections for LTS and MCD. Metrika 55, 111–123.

Hubert, M., Rousseeuw, P. J. and Verdonck, T. (2012) A deterministic algorithm for robust location and scatter. Journal of Computational and Graphical Statistics 21, 618–637.

Examples

n<-100
set.seed(123)# for reproducibility
x0<-matrix(rnorm(n*2),nc=2)
out1<-OGKCStep(x0,alpha=0.5,scale_est=pcaPP::qn)

#comparaison with DetMCD:

#a) create data

set.seed(123456)
Simulation<-DetR:::fx01()
#should be \approx 10
sqrt(min(mahalanobis(Simulation$Data[Simulation$label==0,],rep(0,ncol(Simulation$Data)),
Simulation$Sigma_u))/qchisq(0.975,df=ncol(Simulation$Data)))
a0<-eigen(Simulation$Sigma_u)
Su_ih<-(a0$vector)%*%diag(1/sqrt(a0$values))%*%t(a0$vector)
#run algorithms 
A0<-robustbase::covMcd(Simulation$Data,nsamp='deterministic',scalefn=pcaPP::qn,alpha=0.5)
A1<-OGKCStep(Simulation$Data,alpha=0.5,scale_est=pcaPP::qn)
#getbiases algorithms 
SB<-eigen(Su_ih%*%var(Simulation$Data[A1,])%*%Su_ih)$values
log10(SB[1]/SB[ncol(Simulation$Data)-1])
SB<-eigen(Su_ih%*%var(Simulation$Data[A0$best,])%*%Su_ih)$values
log10(SB[1]/SB[ncol(Simulation$Data)-1])