Publication:
A robust partial least squares method with applications

dc.affiliation.dptoUC3M. Departamento de Estadísticaes
dc.contributor.authorGonzález, Javier
dc.contributor.authorPeña, Daniel
dc.contributor.authorRomera, Rosario
dc.contributor.editorUniversidad Carlos III de Madrid. Departamento de Estadística
dc.date.accessioned2007-03-22T13:53:18Z
dc.date.available2007-03-22T13:53:18Z
dc.date.issued2007-03
dc.description.abstractPartial least squares regression (PLS) is a linear regression technique developed to relate many regressors to one or several response variables. Robust methods are introduced to reduce or remove the effect of outlying data points. In this paper we show that if the sample covariance matrix is properly robustified further robustification of the linear regression steps of the PLS algorithm becomes unnecessary. The robust estimate of the covariance matrix is computed by searching for outliers in univariate projections of the data on a combination of random directions (Stahel-Donoho) and specific directions obtained by maximizing and minimizing the kurtosis coefficient of the projected data, as proposed by Peña and Prieto (2006). It is shown that this procedure is fast to apply and provides better results than other procedures proposed in the literature. Its performance is illustrated by Monte Carlo and by an example, where the algorithm is able to show features of the data which were undetected by previous methods.
dc.format.extent245726 bytes
dc.format.mimetypeapplication/pdf
dc.identifier.repecws071304
dc.identifier.urihttps://hdl.handle.net/10016/665
dc.language.isoeng
dc.language.isoeng
dc.relation.ispartofseriesUC3M Working papers. Statistics and Econometrics
dc.relation.ispartofseries07-04
dc.rightsAtribución-NoComercial-SinDerivadas 3.0 España
dc.rights.accessRightsopen access
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/3.0/es/
dc.subject.ecienciaEstadística
dc.subject.otherKurtosis
dc.subject.otherProjections
dc.subject.otherRobust covariance matrix
dc.subject.otherStahel-Donoho estimator
dc.titleA robust partial least squares method with applications
dc.typeworking paper*
dspace.entity.typePublication
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
ws071304.pdf
Size:
239.97 KB
Format:
Adobe Portable Document Format