Routines classifies codes of missing valuesas numbers in objects of the compositions package.
missingSummary(x,..., vlabs = colnames(x),
mc=attr(x,"missingClassifier"),
values=eval(formals(missingType)$values))
missingType(x,..., mc=attr(x,"missingClassifier"),
values=c("NMV", "BDT", "MAR", "MNAR", "SZ", "Err"))
a dataset which might contain missings
additional arguments for mc
optionally in missingSummary, an alternate routine to be used
instead of missingType
labels for the variables
the names of the different types of missings. "Err"
is a
value that can not be classified e.g. Inf
.
missingType
returns a character vector/matrix with the same dimension and
dimnames as x
giving the type of every value.
missingSummary
returns a table giving the number of missings of each
type for each variable.
The function mainly counts the various types of missing values.
Boogaart, K.G., R. Tolosana-Delgado, M. Bren (2006) Concepts for the handling of zeros and missings in compositional data, Proceedings of IAMG 2006, Liege
# NOT RUN {
data(SimulatedAmounts)
x <- acomp(sa.lognormals)
xnew <- simulateMissings(x,dl=0.05,MAR=0.05,MNAR=0.05,SZ=0.05)
xnew
missingSummary(xnew)
# }
Run the code above in your browser using DataLab