rg.mvalloc: Robust Multivariate Allocation Procedure

In StatDA: Statistical Analysis for Environmental Data

Robust Multivariate Allocation Procedure

Description

Function to allocate an individual to one of several populations.

Usage

rg.mvalloc(pcrit = 0.05, x, ...)

Arguments

pcrit	When the probability of group membership is less than pcrit it is allocated to group 0.
x	contains the individuals to be allocated
...	arguments for creating a list of groups

Details

m objects are the reference populations generated by md.gait, rg.robmva or rg.mva to estimate Mahalanobis distancesand predicted probabilities of group membership for individuals in matrix x. Note that the log |determinant| of the appropriate covariance matrix is added to the Mahalanobis distance on the assumption that the covariance matrices are inhomogeneous. If the data require transformation this must be undertaken before calling this function. This implies that a similar transformation must have been used for all the reference data subsets.

Value

groups	the groups
m	number of groups
n	number of individuals to be allocated
p	number of columns
pgm	number of individuals to be allocated multiplied with the groups
pcrit	critical probability
xalloc	number of individuals as integer

Author(s)

Peter Filzmoser <[e-mail]> http://cstat.tuwien.ac.at/filz/

References

C. Reimann, P. Filzmoser, R.G. Garrett, and R. Dutter: Statistical Data Analysis Explained. Applied Environmental Statistics with R. John Wiley and Sons, Chichester, 2008.

Examples

#input data
data(ohorizon)
vegzn=ohorizon[,"VEG_ZONE"]
veg=rep(NA,nrow(ohorizon))
veg[vegzn=="BOREAL_FOREST"] <- 1
veg[vegzn=="FOREST_TUNDRA"] <- 2
veg[vegzn=="SHRUB_TUNDRA"] <- 3
veg[vegzn=="DWARF_SHRUB_TUNDRA"] <- 3
veg[vegzn=="TUNDRA"] <- 3
el=c("Ag","Al","As","B","Ba","Bi","Ca","Cd","Co","Cu","Fe","K","Mg","Mn",
  "Na","Ni","P","Pb","Rb","S","Sb","Sr","Th","Tl","V","Y","Zn")
x <- log10(ohorizon[!is.na(veg),el])
v <- veg[!is.na(veg)]

res.zone1=rg.mva(as.matrix(x[v==1,]))
res.zone2=rg.mva(as.matrix(x[v==2,]))
res.zone3=rg.mva(as.matrix(x[v==3,]))
res=rg.mvalloc(pcrit=0.01,x,res.zone1,res.zone2,res.zone3)

StatDA documentation built on June 7, 2023, 6:26 p.m.