expit {rje}

Expit and Logit.

Description

Functions to take the expit and logit of numerical vectors.

Usage

expit(x)

logit(x)

Arguments

x

vector of real numbers; for logit to return a sensible value these should be between 0 and 1.

Details

logit implements the usual logit function, which is

logit(x) = \log\frac{x}{1-x},

and expit its inverse:

expit(x) = \frac{e^x}{1+e^x}.

It is assumed that logit(0) = -Inf and logit(1) = Inf, and correspondingly for expit.

Value

A real vector corresponding to the expits or logits of x

Functions

• logit: logit function

Warning

Choosing very large (positive or negative) values to apply to expit may result in inaccurate inversion (see example below).

Author(s)

Robin Evans

Examples

x = c(5, -2, 0.1)
y = expit(x)
logit(y)

# Beware large values!
logit(expit(100))