logistic: Apply a custom logisitic transform to a vector

In bossMaps: Convert Binary Species Range Maps into Continuous Surfaces Based on Distance to Range Edge

Description

Performs a generalized logistic transform (aka Richard's curve) as follows:

y=upper-(upper-lower)/((1+exp(-rate*(x-shift)))^(1/skew))

Usage

logistic(x, lower = 0, upper = 1, rate = 0.04, skew = 0.2, shift = 0,
  parms = NULL)

Arguments

x	vector or raster* object to apply logistic transform.
lower	The lower asymptote
upper	The upper asymptote
rate	The rate of decay. A rate of 0 indicates no decay, while large rates (>10) become like step-functions.
skew	Affects near which asymptote maximum rate (skew must be >0). If skew=1, the curve becomes a standard symetrical logistic centered at m. As skew approaches zero, it becomes more asymetrical.
shift	The time of maximum decay. Slides the curve left-right.
parms	an optional list with each of the following named parameters (e.g. list(lower=0.01,upper=1,rate=0.04,skew=0.2,shift=0)). Use of parms rather than naming each parameter separately is for convenience with the optim function.

References

Richards, F. J. (1959). 'A Flexible Growth Function for Empirical Use'. Journal of Experimental Botany 10 (2): 290-300.

Examples

curve(logistic(x,rate=0.05,skew=0.5),-150,300)

bossMaps documentation built on May 2, 2019, 3:57 p.m.