autoconst {prabclus}

Spatial autocorrelation parameter estimation

Description

Monte Carlo estimation of the disjunction/spatial autocorrelation parameter pd for the simulation model used in randpop.nb, used for tests for clustering of presence-absence data.

autoconst is the main function; autoreg performs the simulation and is executed within autoconst.

Usage

autoconst(x, prange = c(0, 1), twostep = TRUE, step1 = 0.1,
step2 = 0.01, plot = TRUE, nperp = 4, ejprob = NULL,
species.fixed = TRUE, pdfnb=FALSE, ignore.richness=FALSE)

autoreg(x, probs, ejprob, plot = TRUE, nperp = 4, species.fixed = TRUE,
pdfnb=FALSE, ignore.richness=FALSE)

Arguments

x

object of class prab as generated by prabinit. Presence-absence data to be analyzed.

prange

numerical range vector, lower value not smaller than 0, larger value not larger than 1. Range where the parameter is to be found.

twostep

logical. If TRUE, a first estimation step is carried out in the whole prange, and then the final estimation is determined between the preliminary estimator -5*step2 and +5*step2. Else, the first simulation determines the final estimator.

step1

numerical between 0 and 1. Interval length between subsequent choices of pd for the first simulation.

step2

numerical between 0 and 1. Interval length between subsequent choices of pd for the second simulation in case of twostep=TRUE.

plot

logical. If TRUE, a scatterplot of pd-values against resulting ejprob values (see below), with regression line and data value of ejprob is shown.

nperp

integer. Number of simulations per pd-value.

ejprob

numerical between 0 and 1. Observed disjunction probability for data x; if not specified in advance, it will be computed by autoconst.

species.fixed

logical. If TRUE, sizes of generated species match the species sizes in x, else they are generated from the empirical distribution of species sizes in x.

probs

vector of numericals between 0 and 1. pd values for the simulation.

pdfnb

logical. If TRUE, the probabilities of the regions are modified according to the number of neighboring regions in randpop.nb, see Hennig and Hausdorf (2002), p. 5.

ignore.richness

logical. If TRUE, there is no assumption of species richnesses to differ between regions in the null model. Regionwise probabilities don't differ in the generation of null data.

Details

The spatial autocorrelation parameter pd of the model for the generation of presence-absence data sets used by randpop.nb can be estimated by use of the observed disjuction probability ejprob which is the sum of all species' connectivity components minus the number of species divided by the number of "presence" entries minus the number of species. This is done by a simulation of artificial data sets with characteristics of x and different pd-values, governed by prange, step1, step2 and nperp. ejprob is then calculated for all simulated populations. A linear regression of ejprob on pd is performed and the estimator of pd is determined by computing the inverse of the regression function for the ejprob-value of x.

Value

autoconst produces the same list as autoreg with additional component ejprob. The components are

pd

(eventually) estimated parameter pd.

coef

(eventually) estimated regression coefficients.

ejprob

see above.

Author(s)

Christian Hennig [e-mail] https://www.unibo.it/sitoweb/christian.hennig/en

References

Hausdorf, B. and Hennig, C. (2003) Biotic Element Analysis in Biogeography. To appear in Systematic Biology.

Hausdorf, B. and Hennig, C. (2003) Nestedness of north-west European land snail ranges as a consequence of differential immigration from Pleistocene glacial refuges. Oecologia 135, 102-109.

Hennig, C. and Hausdorf, B. (2004) Distance-based parametric bootstrap tests for clustering of species ranges. Computational Statistics and Data Analysis 45, 875-896.

See Also

randpop.nb, prabinit, con.comp

Examples

options(digits=4)
data(kykladspecreg)
data(nb)
set.seed(1234)
x <- prabinit(prabmatrix=kykladspecreg, neighborhood=nb)
ax <- autoconst(x,nperp=2,step1=0.3,twostep=FALSE)