sigmoid_core: Sigmoid curve with gradient

In mnirs: Muscle Near-Infrared Spectroscopy Processing and Analysis

Sigmoid curve with gradient

Description

sigmoid_core() evaluates a 4-parameter sigmoid of the given shape and its partial derivatives on the canonical parameters, shared by the selfStart model functions of SSlogistic(), SSgompertz(), SSgompertz_left(), and SSsigmoidal_drift(). Every shape is a function W(u) of u = k * (t - xmid) with rate k = c * slope / (B - A) (c = 4 symmetric, e Gompertz), so with P = dW/du the partials share one form. sigmoid_model() attaches the gradient over the parameters written as bare symbols in mCall (see free_params()), so stats::nls() skips stats::numericDeriv().

Usage

sigmoid_core(t, A, B, xmid, slope, shape)

sigmoid_model(mCall, t, A, B, xmid, slope, shape)

Arguments

t	A numeric vector of the predictor variable (time).
A	A numeric parameter for the starting asymptote of the response variable.
B	A numeric parameter for the ending asymptote of the response variable.
xmid	A numeric parameter for the time at the inflection point (the steepest point) of the curve, in units of the predictor variable t.
slope	A numeric parameter for the response rate dx/dt at the inflection xmid.
shape	Character; the 4-parameter sigmoidal shape. One of "symmetric" (default; logistic()), "gompertz" (gompertz()), or "gompertz_left" (gompertz_left()).
mCall	A matched call to the model function.

Value

sigmoid_core(): a list of the curve val, the partial derivatives by parameter name, and the rate k. sigmoid_model(): a numeric vector of predicted values with a "gradient" attribute when any parameter is free.

mnirs documentation built on Sept. 13, 2026, 1:06 a.m.