-
Notifications
You must be signed in to change notification settings - Fork 31
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Statistical test used to calculate p-values for Poisson regression #483
Comments
I don't know how to answer this. People in different fields call it different things. For GLMs you typically obtain a z ratio of the observed difference divided by its asymptotic standard error. It is used to test a null hypothesis that the true difference of means is zero versus a two-sided alternative. And by default with comparisons, a Tukey adjustment is applied to the P values, which are thus right-tail areas of the Studentized Range distribution. If you need more clarification , I suggest posting a question on CrossValidated |
Thank you! |
Ok I posted it on CrossValidated, because I also wanted to confirm the default for 'pairs' is a two-sided test. |
The documentation for |
Thanks, but I don't see a default value for 'side' mentioned in that documentation, I just read through it. Maybe I'm missing it somewhere? |
Hi,
I wanted to clarify - what is the correct terminology to describe the statistical test that calculates a p-value for a specific predictor variable in a Poisson regression.
i.e. emmeans(~glm(formula = ...), family=poisson) %>% pairs -> produces p-values. What is the statistical test used to calculate these?
Thanks
The text was updated successfully, but these errors were encountered: