logistic: Analysis: Logistic regression

In AgroR: Experimental Statistics and Graphics for Agricultural Sciences

Analysis: Logistic regression

Description

Logistic regression is a very popular analysis in agrarian sciences, such as in fruit growth curves, seed germination, etc...The logistic function performs the analysis using 3 or 4 parameters of the logistic model, being imported from the LL function .3 or LL.4 of the drc package (Ritz & Ritz, 2016).

Usage

logistic(
  trat,
  resp,
  npar = "LL.3",
  error = "SE",
  ylab = "Dependent",
  xlab = expression("Independent"),
  theme = theme_classic(),
  legend.position = "top",
  r2 = "all",
  width.bar = NA,
  scale = "none",
  textsize = 12,
  font.family = "sans"
)

Arguments

trat	Numerical or complex vector with treatments
resp	Numerical vector containing the response of the experiment.
npar	Number of model parameters
error	Error bar (It can be SE - default, SD or FALSE)
ylab	Variable response name (Accepts the expression() function)
xlab	Treatments name (Accepts the expression() function)
theme	ggplot2 theme (default is theme_bw())
legend.position	Legend position (default is c(0.3,0.8))
r2	Coefficient of determination of the mean or all values (default is all)
width.bar	Bar width
scale	Sets x scale (default is none, can be "log")
textsize	Font size
font.family	Font family (default is sans)

Details

The three-parameter log-logistic function with lower limit 0 is

f(x) = 0 + \frac{d}{1+\exp(b(\log(x)-\log(e)))}

The four-parameter log-logistic function is given by the expression

f(x) = c + \frac{d-c}{1+\exp(b(\log(x)-\log(e)))}

The function is symmetric about the inflection point (e).

Value

The function allows the automatic graph and equation construction of the logistic model, provides important statistics, such as the Akaike (AIC) and Bayesian (BIC) inference criteria, coefficient of determination (r2), square root of the mean error ( RMSE).

Author(s)

Model imported from the drc package (Ritz et al., 2016)

Gabriel Danilo Shimizu

Leandro Simoes Azeredo Goncalves

References

Seber, G. A. F. and Wild, C. J (1989) Nonlinear Regression, New York: Wiley and Sons (p. 330).

Ritz, C.; Strebig, J.C.; Ritz, M.C. Package ‘drc’. Creative Commons: Mountain View, CA, USA, 2016.

Examples

data("emerg")
with(emerg, logistic(time, resp,xlab="Time (days)",ylab="Emergence (%)"))
with(emerg, logistic(time, resp,npar="LL.4",xlab="Time (days)",ylab="Emergence (%)"))

AgroR documentation built on May 29, 2024, 4:18 a.m.