Hello to everybody
I have sent you another dataset than I thougt. The description is still valid but there are more NAs in it so that some factor levels are less populated than in the other dataset.
To get the same "quality" for the dataset you need to execute two lines of code in addition.
library(robustbase)
dat <- na.omit(read.table(file="air.txt"))
jul <- names(table(dat$julday))[table(dat$julday)>=20]
dat <- dat[is.element(dat$julday,jul),]
dat$julday <- factor(dat$julday)
dim(dat) ## 8088 5
r.MM01 <- lmrob(LNOx ~ LNOxEm + sqrtWS + julday,
dat=dat, compute.rd = FALSE)
Kind regards
Ren?
Ren? Locher E-Mail: rene.locher at zhwin.ch
Institut f?r Datenanalyse und Prozessdesign Tel: 052/ 267 7810
Z?rcher Hochschule Winterthur Fax: 052/ 268 7810
Technopark / J?gerstrasse 2
Postfach
CH-8400 Winterthur http://www.idp.zhwin.ch
[RsR] Addendum to test dataset for robustbase V 0.1-5
1 message · Locher René (lor)