Skip to main content

Module exp_link

ndarray_glm

Module exp_link 

Source
Expand description

Link functions for exponential regression

Structsยง

Log
The log link $g(\mu) = \log(\mu)$ avoids linear predictors that give negative expectations.
NegRec
The canonical link function for exponential regression is the negative reciprocal $\eta = -1/mu$. This fails to prevent negative predicted y-values.