Computes functional rarity indices as proposed by Violle et al.
(2017)

`funrar`

is a package to compute functional rarity indices, it quantifies how species are rare both from a functional and an extent point of view. Following the different facets of rarity proposed by Rabinowitz (1981). See this reference for more details on Functional Rarity indices:

**NEWS: the funrar paper just got published**

Please cite the following reference when using `funrar`

in a paper:

Grenié M, Denelle P, Tucker CM, Munoz F, Violle C. funrar: An R package to characterize functional rarity. Divers Distrib. 2017;00:1–7. https://doi.org/10.1111/ddi.12629

or refer to the CITATION file, using:

`citation(package = "funrar")`

The package is on CRAN, you can install it using:

`install.packages("funrar")`

If you want to have the latest development version use `devtools`

:

`# install.packages("devtools") # If 'devtools' is not installed yetdevtools::install_github("Rekyt/funrar", build_vignettes = TRUE)`

Apart from base packages dependencies, `funrar`

depends on `dplyr`

and `cluster`

.

In addition to code example included in help of functions, two vignettes explain how to use the package. The functional rarity indices vignette explains in details the different indices and function provided; while the sparse matrices vignette shows how to use sparse matrices to gain speed in memory when computing functional rarity indices.

Access the vignette through R using the `vignette()`

function.

Rabinowitz D., Seven forms of rarity In The Biological Aspects of Rare Plant Conservation (1981), pp. 205-217

- Fix a bug in the test of
`distinctiveness_dimensions()`

that generated errors on cran server.

- Add Authors' ORCID and all contributors;
- funrar paper is now included in DESCRIPTION, README.md and has a proper CITATIONq file;
- Fix typos in documentation;
- Transformation from tidy data.frame to sparse matrix is now possible using
`stack_to_matrix(x, sparse = TRUE)`

(#25); - Add a warning message when using only continuous traits with function
`comput_dist_matrix()`

, as it defaults to Gower's distance (#27); - Specification in help that functional distances need to be scaled between 0 and 1 prior to distinctiveness computation (#26).

- Split
`rarity_dimensions()`

in two more explicit functions:`uniqueness_dimensions()`

and`distinctiveness_dimensions()`

split corresponding tests; - Add internal function to compute multiple functional distance matrix using a single trait table (
`combination_trait_dist()`

); `distinctiveness()`

now fully conserve the dimnames of the provided site-species matrix.

- Add tests for
`rarity_dimensions()`

; `rarity_dimensions()`

now comprises both Uniqueness and Distinctiveness;- Remove packages
`StatMatch`

,`microbenchmark`

&`reshape2`

from suggested packages.

- Made
`make_absolute()`

defunct because it was based on false assumptions and would not give back matrices of relative abundances; - Improved examples of
`make_relative()`

,`uniqueness()`

,`distinctiveness()`

to compute across single communities or regional pools; - Add
`rarity_dimensions()`

function to measure the different facets of rarity according to the trait; - Add
`center`

and`scale`

arguments in`compute_dist_matrix()`

to scale traits before computing distance, these arguments are sensitive to the specific distance metric used; - Use markdown with
`roxygen2`

to generates documentation.

- Corrected bug so that dense matrices can be transformed to stack data frame using
`matrix_to_stack()`

(#19), - Updated citation for Violle et al. 2017,
- Use package
`goodpractice`

to enforce better code style, - Add
`is_relative()`

function to test if matrix contains relative abundances,`scarcity()`

and`distinctiveness()`

now warns if it is not the case (#21), - Conditionnally use
`microbenchmark`

following CRAN advices.

- Added functions to convert absolute abundance matrix to relative abundance matrix,
`make_relative()`

and reverse function`make_absolute()`

, - Added a
`NEWS.md`

file to track changes to the package.