spatialwarnings (version 3.0.3)

xmin_estim: Estimate the minimum patch size of a power-law distribution

Description

When fitting a power-law to a discrete distribution, it might be worth discarding points below a certain threshold (xmin) to improve the fit. This function estimates the optimal xmin based on the Kolmogorov-Smirnoff distance between the fit and the empirical distribution, as suggested by Clauset et al. (2009).

Usage

xmin_estim(dat, bounds = range(dat))

Value

The estimated xmin as an integer value

Arguments

dat

A vector of integer values

bounds

A vector of two values representing the bounds in which the best xmin is searched

Details

The function returns NA if dat has only three unique values or if the power-law fit failed.

References

Clauset, A., Shalizi, C. R., & Newman, M. E. (2009). Power-law distributions in empirical data. SIAM review, 51(4), 661-703.

See Also

patchdistr_sews

Examples

if (FALSE) { 
psd <- patchsizes(forestgap[[5]])
xmin_estim(psd)
}