favClass: Classify favourability into 3 categories (low, intermediate,...

In fuzzySim: Fuzzy Similarity in Species Distributions

Classify favourability into 3 categories (low, intermediate, high)

Description

This function takes a vector of Favourability values and reclassifies them into 3 increasing categories: low, intermediate or high. By default, the breaks between these classes are 0.2 and 0.8 (see Details), although these can be changed by the user.

Usage

favClass(fav, breaks = c(0.2, 0.8), character = FALSE)

Arguments

fav	a numeric vector of favourability values (obtained, e.g., with functions Fav or multGLM).
breaks	a numeric vector of length 2 containing the two values which will divide fav into the 3 classes. Defaults to c(0.2, 0.8) following the literature (see Details).
character	logical value indicating whether the result should be returned in character rather numeric form. Defaults to FALSE.

Details

Some applications of species distribution models imply setting a threshold to separate areas with high and low probability or favourability for occurrence (see, e.g., bioThreat). However, it makes little sense to establish as markedly different areas with, for example, 0.49 and 0.51 favourability values (Hosmer & Lemeshow, 1989). It may thus be wiser to open a gap between values considered as clearly favourable and clearly unfavourable. When this option is taken in the literature, commonly used breaks are 0.8 as a threshold to classify highly favourable values, as the odds are more than 4:1 favourable to the species; 0.2 as a threshold below which to consider highly unfavourable values, as odds are less than 1:4; and classifying the remaining values as intermediate favourability (e.g., Munoz & Real 2006, Olivero et al. 2016).

Value

This function returns either an integer or a character vector (following the 'character' argument, which is set to FALSE by default), of the same length as fav, reclassifying it into 3 categories: 1 ('low'), 2 ('intermediate'), or 3 ('high').

Author(s)

A. Marcia Barbosa

References

Hosmer D.W. Jr & Lemeshow S. (1989) Applied logistic regression. John Wiley & Sons, New York

Munoz A.R. & Real R. (2006) Assessing the potential range expansion of the exotic monk parakeet in Spain. Diversity and Distributions, 12: 656-665

Olivero J., Fa J.E., Real R., Farfan M.A., Marquez A.L., Vargas J.M., Gonzalez J.P., Cunningham A.A. & Nasi R. (2017) Mammalian biogeography and the Ebola virus in Africa. Mammal Review, 47: 24-37

See Also

Fav, multGLM

Examples

data(rotif.env)
mods <- multGLM(rotif.env, sp.cols = 20, var.cols = 5:17)
fav <- mods$predictions[ , 2]
data.frame(fav = fav, favcl_num = favClass(fav), 
favcl_chr = favClass(fav, character = TRUE))

fuzzySim documentation built on Oct. 9, 2023, 5:09 p.m.