Generalized Logit and Inverse Logit Function
Compute generalized logit and generalized inverse logit functions.
Logit(x, min = 0, max = 1) LogitInv(x, min = 0, max = 1)
x |
value(s) to be transformed |
min |
lower end of logit interval |
max |
upper end of logit interval |
The generalized logit function takes values on [min, max] and
transforms them to span [-Inf, Inf].
It is defined as:
y = log(p/(1-p)) where p=(x-min)/(max-min)
The generalized inverse logit function provides the inverse transformation:
x = p' * (max-min) + min where p' = exp(y)/(1+exp(y))
Transformed value(s).
Gregory R. Warnes greg@warnes.net
x <- seq(0,10, by=0.25) xt <- Logit(x, min=0, max=10) cbind(x,xt) y <- LogitInv(xt, min=0, max=10) cbind(x, xt, y)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.