TITEL : Antille/El May Example KEYWORDS (statist) : robust regression, outlier ANWENDUNGSGEBIET : -- VERW.(VORL./J.-S.) : -- Ueb.aufgabe : " .tex" -------------------------------------------------------------------------------- KURZE BESCHREIBUNG : Künstliche Daten (!!) ~~~~~~~~~~ 1 Ausreisser in 12 Beobachtungen; trotzdem brechen LMS und LTS (& mmreg) zusammen. M. Maechler hat dies aus Y.Dodge "Thoughts on Real Data and Statistics", on p.97 in @Book{HinHF92, author = {H. Hinterberger and J. French}, title = {Proceedings of the Sixth International Working Conference on Scientific and Statistical Database Management}, publisher = ETH, year = 1992 } where the data is cited from @inproceedings{AntGM92, author = {Gerard Antille and May, H. El}, title = {The Use of Slices in the {LMS} and the Method of Density Slices: {F}oundation and Comparison}, year = 1992, editor = {Yadolah Dodge and Joe Whittaker}, booktitle = {COMPSTAT: Proc.\ 10th Symp.\ Computat.\ Statist., Neuchatel}, ETHbib = {P 714 301: 10}, ISBN = {3-7908-0634-X (Physica). 0-387-91429-3 (Springer)}, publisher = {Physica-Verlag}, keywords = {Regression; Classification; Least median of squares}, pages = {441--445}, volume = 1, } --> /u/maechler/R/MM/STATISTICS/robust/regression/AMay-ex.S tut einiges mit diesen Daten SPEZ. EIG. : 1 Hebel-Ausreisser; Punkte nicht in allgemeiner Lage. -------------------------------------------------------------------------------- DIM.(VAR./FAELLE) : 2 / 12 Einlesen mit S-plus: read.table("/u/sfs/ueb/datasets/antille-may-ex.dat",head=T) -------------------------------------------------------------------------------- * / LITERATUR : / s. oben * D=Daten, vA=volle Analyse, tA=teilweise Analyse, K=Kommentar -------------------------------------------------------------------------------- BEMERKUNGEN : -------------------------------------------------------------------------------- USERS : DATUM/UNTERSCHRIFT : Jan.1996 / M.Maechler