R: Constructor function for objects of class RobEstControl

RobEstControl {MAINT.Data}

R Documentation

Constructor function for objects of class RobEstControl

Description

This function will create a control object of class RobEstControl containing the control parameters for the robust estimation functions fasttle, RobMxtDEst, Roblda and Robqda.

Usage

RobEstControl(alpha=0.75, nsamp=500,  seed=NULL, trace=FALSE, use.correction=TRUE,
  ncsteps=200, getalpha="TwoStep", rawMD2Dist="ChiSq", MD2Dist="ChiSq", eta=0.025,
  multiCmpCor="never",  getkdblstar="Twopplusone", outlin="MidPandLogR", 
  trialmethod="simple", m=1, reweighted=TRUE, k2max=1e6, otpType="SetMD2andEst")

Arguments

`alpha`	Numeric parameter controlling the size of the subsets over which the trimmed likelihood is maximized; roughly alpha*nrow(Sdt) observations are used for computing the trimmed likelihood. Allowed values are between 0.5 and 1. Note that when argument ‘getalpha’ is set to “TwoStep” the final value of ‘alpha’ is estimated by a two-step procedure and the value of argument ‘alpha’ is only used to specify the size of the samples used in the first step.
`nsamp`	Number of subsets used for initial estimates.
`seed`	Starting value for random generator.
`trace`	Whether to print intermediate results.
`use.correction`	Whether to use finite sample correction factors.
`ncsteps`	The maximum number of concentration steps used each iteration of the fasttle algorithm.
`getalpha`	Argument specifying if the ‘alpha’ parameter (roughly the percentage of the sample used for computing the trimmed likelihood) should be estimadted from the data, or if the value of the argument ‘alpha’ should be used instead. When set to “TwoStep”, ‘alpha’ is estimated by a two-step procedure with the value of argument ‘alpha’ specifying the size of the samples used in the first step. Otherwise the value of argument ‘alpha’ is used directly.
`rawMD2Dist`	The assumed reference distribution of the raw MCD squared distances, which is used to find to cutoffs defining the observations kept in one-step reweighted MCD estimates. Alternatives are ‘ChiSq’,‘HardRockeAsF’ and ‘HardRockeAdjF’, respectivelly for the usual Chi-squared, and the asymptotic and adjusted scaled F distributions proposed by Hardin and Rocke (2005).
`MD2Dist`	The assumed reference distributions used to find cutoffs defining the observations assumed as outliers. Alternatives are “ChiSq” and “CerioliBetaF” respectivelly for the usual Chi-squared, the Beta and F distributions proposed by Cerioli (2010).
`eta`	Nominal size of the null hypothesis that a given observation is not an outlier. Defines the raw MCD Mahalanobis distances cutoff used to choose the observations kept in the reweightening step.
`multiCmpCor`	Whether a multicomparison correction of the nominal size (eta) for the outliers tests should be performed. Alternatives are: ‘never’ – ignoring the multicomparisons and testing all entities at ‘eta’. ‘always’ – testing all n entitites at 1.- (1.-‘eta’^(1/n)); and ‘iterstep’ – as sugested by Cerioli (2010), make an initial set of tests using the nominal size 1.- (1-‘eta’^(1/n)), and if no outliers were detected stop. Otherwise, make a second step testing for outliers at ‘eta’.
`getkdblstar`	Argument specifying the size of the initial small (in order to minimize the probability of outliers) subsets. If set to the string “Twopplusone” (default) the initial sets have twice the number of interval-value variables plus one which are they are the smaller samples that lead to a non-singular covaraince estimate). Otherwise, an integer with the size of the initial sets.
`outlin`	The type of outliers to be considered. “MidPandLogR” if outliers may be present in both MidPpoints and LogRanges, “MidP” if outliers are only present in MidPpoints, or “LogR” if outliers are only present in LogRanges.
`trialmethod`	The method to find a trial subset used to initialize each replication of the fasttle algorithm. The current options are “simple” (default) that simply selects ‘kdblstar’ observations at random, and “Poolm” that divides the original sample into ‘m’ non-overlaping subsets, applies the ‘simple trial’ and the refinement methods to each one of them, and merges the results into a trial subset.
`m`	Number of non-overlaping subsets used by the trial method when the argument of ‘trialmethod’ is set to 'Poolm'.
`reweighted`	Should a (Re)weighted estimate of the covariance matrix be used in the computation of the trimmed likelihood or just a “raw” covariance estimate; default is (Re)weighting.
`k2max`	Maximal allowed l2-norm condition number for correlation matrices. Correlation matrices with condition number above k2max are considered to be numerically singular, leading to degenerate results.
`otpType`	The amount of output returned by fasttle. Current options are “SetMD2andEst” (default) which returns an ‘IdtSngNDRE’ object with the fasttle estimates, a vector with the final trimmed subset elements used to compute these estimates and the corresponding robust squared Mahalanobis distances, and “SetMD2EstandPrfSt” wich returns an ‘IdtSngNDRE’ object with the previous slots plust a list of some performance statistics concerning the algorithm execution.

Value

A RobEstControl object

References

Brito, P., Duarte Silva, A. P. (2012): "Modelling Interval Data with Normal and Skew-Normal Distributions". Journal of Applied Statistics, Volume 39, Issue 1, 3-20.

Cerioli, A. (2010), Multivariate Outlier Detection with High-Breakdown Estimators. Journal of the American Statistical Association 105 (489), 147–156.

Duarte Silva, A.P., Filzmoser, P. and Brito, P. (2017), Outlier detection in interval data. Advances in Data Analysis and Classification, 1–38.

Hardin, J. and Rocke, A. (2005), The Distribution of Robust Distances. Journal of Computational and Graphical Statistics 14, 910–927.