radpred: Predicted ranked abundance of species

In sads: Maximum Likelihood Models for Species Abundance Distributions

Predicted ranked abundance of species

Description

Creates an object of rad-class with the ranked abundances predicted by a species abundance distribution or a rank-abundance distribution.

Arguments

object	an object of class fitrad or fitsad; fitted model of rank-abundance or species abundances distributions. Alternatively a numeric vector abundances of species.
sad, rad	character; root name of sad or rad distribution to calculate expected percentiles. See fitsad and fitrad for available distributions.
coef	named list of numeric values; parameter values of the distribution given in sad or rad. Parameters should be named as in the corresponding density function, and in the same order.
trunc	non-negative integer, trunc > min(x); truncation point if fitted distribution is truncated.
distr	Deprecated since sads 0.2.4. See distr function
S	positive integer; number of species in the sample.
N	positive integer; number of individuals in the sample.

Methods

signature(object = "fitrad", sad = "missing", rad = "missing", coef = "missing", trunc = "missing", distr = "missing", S = "missing", N = "missing")

ranked abundances of species predicted from a rank-abundance model fitted with function fitrad.

signature(object = "fitsad", sad = "missing", rad = "missing", coef = "missing", trunc = "missing", distr = "missing", S = "missing", N = "missing")

ranked abundances of species predicted from a abundance distribution model fitted with function fitsad.

signature(object = "fitsadC", sad = "missing", rad = "missing", coef = "missing", trunc = "missing", distr = "missing", S = "missing", N = "missing")

ranked abundances of species predicted from a abundance distribution model fitted with function fitsadC.

signature(object = "missing", sad = "character", rad = "missing", coef = "list", trunc = "ANY", distr = "ANY", S = "numeric", N = "numeric")

ranked abundances of species predicted from abundance distribution named by sad with parameters defined in coef.

signature(object = "numeric", sad = "character", rad = "missing", coef = "list", trunc = "ANY", distr = "ANY", S = "missing", N = "missing")

same as previous method, but with S and N taken from a vector of abundances given by object.

signature(object = "missing", sad = "missing", rad = "character", coef = "list", trunc = "ANY", distr = "missing", S = "numeric", N = "numeric")

ranked abundances of species predicted from a rank-abundance distribution named by rad with parameters defined in coef.

signature(object = "numeric", sad = "missing", rad = "character", coef = "list", trunc = "ANY", distr = "missing", S = "missing", N = "missing")

same as previous method, but with S and N taken from a vector of abundances given by object.

Note

The rank-abundance function is the inverse of the quantile function of the species abundance distribution (May 1975). radpred uses numeric interpolation with approxfun to find quantiles, instead of the slower bisection method used in many quantile functions in the package sads. Hence results from radpred and from the quantile functions may not match exactly.

Author(s)

Paulo I. Prado [e-mail] and Murilo Dantas Miranda

References

May, R.M. 1975. Patterns of Species Abundance and Diversity. In Cody, M.L. and Diamond, J.M. (Eds) Ecology and Evolution of Communities. Harvard University Press. pp 81–120.

Examples

## Predicted frequencies from a fitted model
## meta-community zero-sum multinomial for BCI data
moths.mzsm <- fitsad(moths, "mzsm")
moths.mzsm.r <- radpred(moths.mzsm)
## Rank-abundance plot with observed and predicted frequencies
plot(rad(moths))
lines(moths.mzsm.r)
## Alternative model: local zero-sum multinomial
## Alonso & Mckane (Ecol. Lett. 2004, table 1) give theta = 41 and m = 0.77
moths.lzsm.r <- 
  radpred( moths, sad = "volkov",  
           coef =list(theta = 41, m = 0.77, J=sum(moths)) 
         )
## Adding predicted frequencies to the plot
lines(moths.lzsm.r, col = "red")

sads documentation built on June 22, 2024, 12:18 p.m.