predict_amean_byelt_LOO: Arithmetic mean (amean) by elements occurring within the...

In BenitoJaillard/combinAna: Element clustering that best accounts for an assembly property or function

Description

The numeric vector fct get together the properties of assemblages belonging to different assembly motifs. The properties fct of assemblages belonging to a given assembly motif and containing a given element are separately averaged. The property of each assemblage is computed as the average of mean values of assemblages containing the same elements as the considered assemblage, except the assemblage to predict (leave-one-out, LOO). The motif of each vector element is specified in the vector assMotifs. The elemental composition of each assemblage is specified in the binary matrix mOccur: 0 if the element does not occur, 1 if the element occurs.

Usage

predict_amean_byelt_LOO(fct, assMotifs, mOccur)

Arguments

fct	a vector of numeric values (assembly properties).
assMotifs	a vector of labels of length(fct) (assembly motifs).
mOccur	a matrix of occurrence (occurrence of elements). Its first dimension equals to length(fct). Its second dimension equals to the number of elements.

Details

Prediction is computed using arithmetic mean amean by element byelt occurring within the assemblage and other assemblages of a same motif, by including all assemblages belonging to a same motif, except the assemblage to predict.

Value

Return a vector of same length as fctMot, of which values are computed as an arithmetic mean.

BenitoJaillard/combinAna documentation built on May 9, 2019, 2:17 a.m.