How to find the odds ratios for a logistic model?

Question

I have a standard logistic regression model in R

reg <- glm(formula = y ~ x, family = "binomial"(link='logit'))

I am trying to find the odds ratios for my model in R. Is there a function or some other way to do this?

The title of this question has to do with reporting. The body of this question seems to be about whether there's a good function or way to calculate odds ratios of a model. — Galen, Apr 29 '20 at 17:44
@Galen sorry for the confusion... the title has been updated — 324, Apr 29 '20 at 17:46
This post seems useful: https://stats.stackexchange.com/questions/8661/logistic-regression-in-r-odds-ratio — Anders Ellern Bilgrau, Apr 29 '20 at 17:48

score 6 · Answer 1 · edited Apr 30 '20 at 14:53

6

This is a good way to find the odds ratios

exp(coef(reg))

edited Apr 30 '20 at 14:53

TylerH

answered Apr 29 '20 at 17:51

score 3 · Accepted Answer · answered Apr 29 '20 at 17:53

The log odds ratio can be found by

reg$coefficients

... and the odds ratio would be

exp(reg$coefficients)

... the log of 2.5% and 97.5% levels of the confidence intervals would be

confint.default(reg)

... following that the 2.5% and 97.% levels of the confidence intervals would be

exp(confint.default(reg))

2 Answers2