In the help of the glm command in R, I read

the binomial family the links logit, probit, cauchit, (corresponding to logistic, normal and Cauchy CDFs respectively) log and cloglog
(complementary log-log)

What are the difference between the "normal", Cauchy, log, cloglog models? When should I use them? Are these ONLY referring to different the error models? Anyone can provide Examples or recommendation to prefer one over the other?

  • $\begingroup$ I'm asking exactly about those families not covered in the "Duplicated" you found. I only found some references to the cloglog function, but nothing about the other families... $\endgroup$
    – Diego
    Sep 27, 2017 at 10:35


Browse other questions tagged or ask your own question.