CF {adiv}

Matrices of Intra- and Inter-Specific Similarities

Description

The function CFprop calculates the matrix CF of intra- (diagonal) and inter-specific (off-diagonal) similarities as defined in the main text of Pavoine and Izsak (2014), and matrix CwF as defined in Appendix S3 of Pavoine and Izsak (2014) for weighting functional attributes. The function multiCFprop calculates matrices CwmF1, CwmF2, CwmF3 when several functional traits are considered (Appendix S3 of Pavoine and Izsak 2014). Traits and the attributes of the traits can be weighted. These two functions consider functional traits expressed as proportion (compositional) vectors. The functions CFbinary and multiCFbinary are the equivalents of CFprop and multiCFprop when traits are expressed as binary vectors as shown in Appendix S3 of Pavoine and Izsak (2014).

Usage

CFbinary(df, wA = rep(1, ncol(df)))

multiCFbinary(Ktab, w.attributes = lapply(Ktab, function(x) rep(1, ncol(x))), 
w.traits = rep(1/length(Ktab), length(Ktab)), 
labels = rownames(Ktab[[1]]), solution = c(2, 1))

CFprop(df, wA = rep(1, ncol(df)))

multiCFprop(Ktab, w.attributes = lapply(Ktab, function(x) rep(1, ncol(x))), 
w.traits = rep(1/length(Ktab), length(Ktab)), 
labels = rownames(Ktab[[1]]), solution = c(2, 1))

Arguments

df

a data frame or a matrix with species (or any entities of interest) as rows, functional attributes as columns, and proportions (for CFprop) or 0/1 values (for CFbinary) as entries. The sum of the columns will be standardized to equal 1 in function CFprop but not in CFbinary.

wA

a vector of weights that should be given to the attributes (same order as the columns of df).

Ktab

a list of data frames, each of which represents a trait. For a given trait, the data frame should have species (or any entities of interest) as rows, functional attributes as columns, and proportions (for multiCFprop) or 0/1 values (for multiCFbinary) as entries. The sum of the columns will be standardized to equal 1 in function multiCFprop but not in multiCFbinary.

w.attributes

a list of weights that should be given to the attributes of each trait. Traits should be in the same order as they appear in the list of tables Ktab. The attributes of a trait should be ordered as the columns of the corresponding data frame in Ktab.

w.traits

a numeric vector of weights that should be given to the traits (same order as the tables of Ktab).

labels

a vector of strings that gives the names of the species (or the other entities of interest; same order as the rows of all tables in Ktab).

solution

a numeric value (either 1 or 2) that indicates which equations are used to summarize the information given by several traits among the 2 approaches given in Appendix S3 of Pavoine and Izsak (2014) page 9. If a vector is given, only the first value of the vector is considered.

Value

A matrix with nonnegative values

Author(s)

Sandrine Pavoine [e-mail]

References

Pavoine, S. and Izsak, J. (2014) New biodiversity measure that includes consistent interspecific and intraspecific components. Methods in Ecology and Evolution, 5, 165–172.

See Also

qHdiv