Skip to contents

Inverse Logit (Logistic) Transformation

Source: R/logit_inverse.R
inv_logit.Rd

Applies the inverse logit (logistic) transformation to a real-valued input, mapping it from the real line (-Inf, Inf) to the interval (0, 1).

Usage

inv_logit(x)

Arguments

x

A numeric vector of real values.

Value

A numeric vector where each value is the inverse logit of the corresponding input, representing a probability in the range (0, 1).

Details

The inverse logit function is defined as: $$inv\_logit(x) = \frac{e^x}{1 + e^x}$$ This is often used to map real-valued predictions (e.g., from logistic regression) back to probabilities.

See also

logit for the logit transformation function.

Examples

inv_logit(0)     # Returns 0.5
#> [1] 0.5
inv_logit(1.39)  # Returns approximately 0.8
#> [1] 0.8005922
inv_logit(-1.39) # Returns approximately 0.2
#> [1] 0.1994078