compositions (version 2.0-4)

IsMahalanobisOutlier: Checking for outliers

Description

Detect outliers with respect to a normal distribution model.

Usage

IsMahalanobisOutlier(X,...,alpha=0.05,goodOnly=NULL,
                 replicates=1000,corrected=TRUE,robust=TRUE,crit=NULL)

Value

A logical vector giving for each element the result of the alpha-level test for beeing an outlier. TRUE corresponds to a significant result.

Arguments

X

a dataset (e.g. given as acomp, rcomp, aplus, rplus or rmult) object to which idt and MahalanobisDist apply.

...

further arguments to MahalanobisDist/gsi.mahOutlier

alpha

The confidence level for identifying outliers.

goodOnly

an integer vector. Only the specified index of the dataset should be used for estimation of the outlier criteria. This parameter if only a small portion of the dataset is reliable.

replicates

The number of replicates to be used in the Monte Carlo simulations for determination of the quantiles. The replicates not given a minimum is computed from the alpha level to ensure reasonable precission.

corrected

logical. Literatur often proposed to compare the Mahalanobis distances with Chisq-Approximations of there distributions. However this does not correct for multiple testing. If corrected is true a correction for multiple testing is used. In any case we do not use the chisq-approximation, but a simulation based procedure to compute confidence bounds.

robust

A robustness description as define in robustnessInCompositions

crit

The critical value to be used. Typically the routine is called mainly for the purpose of finding this value, which it does, when crit is NULL, however sometimes we might want to specifiy a value used by someone else to reproduce the results.

Author

K.Gerald v.d. Boogaart http://www.stat.boogaart.de

Details

See outliersInCompositions and robustnessInCompositions for a comprehensive introduction into the outlier treatment in compositions.

See OutlierClassifier1 for a highlevel method to classify observations in the context of outliers.

See Also

OutlierClassifier1 , outlierplot, ClusterFinder1

Examples

if (FALSE) {
data(SimulatedAmounts)

datas <- list(data1=sa.outliers1,data2=sa.outliers2,data3=sa.outliers3,
              data4=sa.outliers4,data5=sa.outliers5,data6=sa.outliers6)

opar<-par(mfrow=c(2,3),pch=19,mar=c(3,2,2,1))
tmp<-mapply(function(x,y){
plot(x,col=ifelse(IsMahalanobisOutlier(x),"red","gray"))
  title(y)
},datas,names(datas))
}