Take a vector fobs
of performances of assemblages
that share a same assembly motif,
and return a vector of performances
predicted as the geometric mean
of performances of assemblages
that contain the same elements as the assemblage to predict.
All assemblages share the same assembly motif.
validate_gmean_byelt_LOO(fobs, assMotif, mOccur)a numeric vector. The vector fobs contains the
quantitative performances of assemblages.
a vector of labels of length(fobs).
The vector assMotif contains the assembly motifs of assemblages.
a matrix of occurrence (occurrence of elements).
Its first dimension equals to length(fobs). Its second dimension
equals to the number of elements.
Return a vector of length(fobs).
Its values are computed as the average
of mean performances of all assemblages
that contain the same elements
as the assemblage to predict (opt.model = "byelt") .
Predicted performances are computed
using geometric mean (opt.mean = "gmean")
of performances.
Assemblages share a same assembly motif (opt.model = "bymot").
Predicted performances are the average
of mean performances of assemblages
that contain the same elements
as the assemblage to predict,
except the assemblage to predict.
This procedure corresponds to a linear model with each assembly motif
based on the element occurrence in each assemblage.
validate_amean_bymot_LOO arithmetic mean
by assembly motif
validate_amean_byelt_LOO arithmetic mean
by elements occurring within assembly motif
validate_gmean_bymot_LOO geometric mean
by assembly motif