fonctional_div_species: Troll Model Validation : functional diversity with...
In Riodinino/TROLLmsc: TROLL Master Thesis Analyses

Troll Model Validation : functional diversity with Leinster-Cobbold Entropy using resampling and "rolled under the armpits" Goodness of Fit statistics.

1
2
3

fonctional_div_species(data, trollsim, speciescol, speciestraits, LMA = "LMA",
  N = "N", P = "P", wsg = "wsg", ah = "ah", hmax = "hmax",
  dmax = "dmax", n_resample = 10000, path, nameplot)

`data`	Data.frame ; An inventory which contains at least species name.
`trollsim`	Data.frame (yet not adapted to trollsim S4 objects)with species name Diameters should correspon in terms of lower bound (generally 10cm) between field data and trollsim : pre-filtering is recommended to make consistent comparisons.
`speciescol`	Character indicating the name of the species identity column
`speciestraits`	Data.frame the contains species name and the 7 traits (species-specific mean values)
`LMA`	Character indicating the name of the LMA species means column
`N`	Character indicating the name of the N species means column
`P`	Character indicating the name of the P species means column
`wsg`	Character indicating the name of the wsg species means column
`ah`	Character indicating the name of the ah species means column
`hmax`	Character indicating the name of the hmax species means column
`dmax`	Character indicating the name of the dmax species means column
`n_resample`	Integer ; the number of times you wish to resample to infer sampling effects.
`path`	Character ; The path to the directory you want to save the graphical outputs into.
`nameplot`	Character ; the name of your plot (used to generate outputs' filenames)

A matrix of computed distances of the resampled Leinster-Cobbold Entropy profiles (LCEP) and the original dataset's LCEP using two distance statistics : (is the first one alright ??) dist = sqrt(sum((LC_data - LC_sample)^2)/nrow(LC_data)) and chi =(sum((LC_sample-LC_data)^2)/LC_data)

Riodinino/TROLLmsc documentation built on May 14, 2019, 3:08 a.m.