rrcov (version 1.5-2)

OsloTransect: Oslo Transect Data

Description

The oslo Transect data set contains 360 samples of different plant species collected along a 120 km transect running through the city of Oslo, Norway.

Usage

data(OsloTransect)

Format

A data frame with 360 observations on the following 38 variables.

X.ID

a numeric vector, unique ID of the sample

X.MAT

a factor with levels BBA BIL BWO FER MOS ROG SNE STW TWI

XCOO

a numeric vector, X coordinate

YCOO

a numeric vector, Y coordinate

XCOO_km

a numeric vector

YCOO_km

a numeric vector

X.FOREST

a factor with levels BIRSPR MIXDEC PINE SPRBIR SPRPIN SPRUCE

DAY

a numeric vector

X.WEATHER

a factor with levels CLOUD MOIST NICE RAIN

ALT

a numeric vector

X.ASP

a factor with levels E FLAT N NE NW S SE SW W

X.GRVEG

a factor with levels BLGR BLLY BLMOLI BLUE BLUGRA GRAS GRBLU GRFE GRMO LYLI MIX MOGR MOSS

X.FLITHO

a factor with levels CAMSED GNEID_O GNEIS_O GNEIS_R MAGM MICSH

Ag_ppb

a numeric vector

As_ash

a numeric vector

B

a numeric vector

Ba

a numeric vector

Ca

a numeric vector

Cd

a numeric vector

Co

a numeric vector

Cr

a numeric vector

Cu

a numeric vector

Fe

a numeric vector

Hg_ppb

a numeric vector

K

a numeric vector

La

a numeric vector

LOI

a numeric vector

Mg

a numeric vector

Mn

a numeric vector

Mo

a numeric vector

Ni

a numeric vector

P

a numeric vector

Pb

a numeric vector

S

a numeric vector

Sb

a numeric vector

Sr

a numeric vector

Ti

a numeric vector

Zn

a numeric vector

Details

Samples of different plant species were collected along a 120 km transect running through the city of Oslo, Norway (forty samples each of leaves, needles,roots or barks of several plant species), and the concentrations of 25 chemical elements for the sample materials are reported. The factors that influenced the observed element concentrations in the sample materials were investigated. This data set was used in Todorov and Filzmoser (2007) for illustration of the robust statistics for one-way MANOVA implemented in the function Wilks.test.

References

Todorov V. and Filzmoser P. (2007) Robust statistic for the one-way MANOVA, submetted to the Journal of Environmetrics.

Examples

# NOT RUN {
data(OsloTransect)
str(OsloTransect)

##
##  Log-transform the numerical part of the data, 
##  choose the desired groups and variables and 
##  perform the classical Wilks' Lambda test
##
OsloTransect[,14:38] <- log(OsloTransect[,14:38])
grp <- OsloTransect$X.FLITHO
ind <- which(grp =="CAMSED" | grp == "GNEIS_O" |
    grp == "GNEIS_R" | grp=="MAGM")
(cwl <- Wilks.test(X.FLITHO~K+P+Zn+Cu,data=OsloTransect[ind,]))

##
## Perform now the robust MCD based Wilks' Lambda test. 
##  Use the already computed multiplication factor 'xd' and 
##  degrees of freedom 'xq' for the approximate distribution.
##

xd <- -0.003708238
xq <- 11.79073
(mcdwl <- Wilks.test(X.FLITHO~K+P+Zn+Cu,data=OsloTransect[ind,], 
    method="mcd", xd=xd, xq=xq))
# }