fitlogdr: Fit dose-response when dose is log10 scale

In TJMurphy/nlfitr: Fitting models to nonlinear data

Description

fitlogdr uses the nlsLM function of the minpack.lm package to fit a model formula to nonlinear response data. The formula is based upon the general hyperbolic function y= ymax/(1+10^((logk-x)*h)), where ymax = yhi - ylo. Both x and logk are in base 10 log scale units. logk, ylo, yhi and h are start list parameters #' for nlsLM. The formula is: 'y = ylo + (yhi - ylo)/(1 + 10^((logk - x)*h))'

Usage

fitlogdr(x, y, data, logk, ylo, yhi, h, weigh)

Arguments

x	a vector of base 10 log scale values, usually dose or concentration.
y	a vector of y values for x, usually responses.
data	a dataframe with x and y.
logk	estimates the value of base 10 log scale x that yields y/ymax = 0.5, usually ED50 or EC50.
ylo	estimates the lowest asymptotic y value, in response units.
yhi	estimates highest asymptotic value, in response units.
h	the Hill slope, a unitless slope factor; -1 > h > 1 is steeper, -1 < h < 1 is shallower. Use negative value for downward sloping response.
weigh	chooses regression weighting by 1/y^2. Default = FALSE.

Value

nls

Examples

# aorta contraction by phenylephrine doses
fitlogdr(x=logM_PE, y=mN, data=aoc, logk=-8, ylo=0, yhi=50, h=1, weigh=TRUE)

# or
fitlogdr(x=lognM_PE, y=mN, data=aoc, logk=1, ylo=0, yhi=50, h=1, weigh=TRUE)

TJMurphy/nlfitr documentation built on March 18, 2021, 12:33 p.m.