boot.coefs.decay {betapart}

Bootstrapping the parameters of distance-decay models computed with decay.model()

Description

Takes the output of decay.model() and bootstraps the parameters of the model (i.e. intercept and slope in negative exponential or power law models, or position parameter and slope in Gompertz models).

Usage

boot.coefs.decay(m1, resamples, st.val = c(1, 0))

Arguments

m1

the output of decay.model().

resamples

the number of bootstrap resamples.

st.val

starting values for the nonlinear model.

Value

The function returns a list with:

model.type

functional form of the model, either negative exponential or power law.

y.type

similarties or dissimilarities.

boot.coefs

a matrix with the coefficients bootstrapped distributions, including values of the first parameter (intercept or position parameter) in the first column, and values of the second parameter (slope) in the second column.

original.coefs

model coefficients as estimated with a nonlinear model using decay.model().

mean.boot

the mean of the bootstrapped distributions.

sd.boot

the standard deviation of the bootstrapped distributions.

Author(s)

Sara Martínez-Santalla, Ramiro Martín-Devasa, Carola Gómez-Rodríguez, Rosa M. Crujeiras, Andrés Baselga

References

Gómez-Rodríguez, C. & Baselga, A. 2018. Variation among European beetle taxa in patterns of distance decay of similarity suggests a major role of dispersal processes. Ecography 41: 1825-1834

Martínez-Santalla S, Martín-Devasa R, Gómez-Rodríguez C, Crujeiras RM, Baselga A. 2022. Assessing the non-linear decay of community similarity: permutation and site-block resampling significance tests. Journal of Biogeography 49: 968-978

See Also

decay.model, zdep

Examples

# presence/absence tables for longhorn beetles of South Europe
data(ceram.s)

# spatial coordinates of territories in South Europe
data(coords.s)

# dissimilarity matrix
ceram.s.sim<-beta.pair(ceram.s)$beta.sim

# spatial distances in km
distgeo.s<-dist(coords.s[,1:2])

# Negative exponential model for the decay of similarity with spatial distance
decay.south<-decay.model(y=1-ceram.s.sim, x=distgeo.s, y.type="sim", model.type="exp")

# Site-block bootstrap
boot.coefs.decay(decay.south, resamples=100)