I am conducting GLM on my count data with Poisson distribution settings in R and I was wondering if I could calculate log odds ratios in the models selected. I can't find it anywhere in the literature or relevant codes for the same. They all point towards binomial distributions. I would appreciate the help.
In Poisson distribution, it is common to exponentiate your coefficients. This is not a log odds ratio, but sometimes called incidence rate ratio. Interpreting the exponentiated values for the Poisson outcome is straightforward.
I agree with Endale Alemayehu Ali, we cannot calculate log odds ratio for count outcome data. We usually calculate log odds ratio for binary outcome data.
Hello Zaara Kidwai,
Like the others, I'm struggling to see the relevance of odds ratios for count data.
Could it be that the counts have a limited range? (e.g. 0 to 5) If that is the case, one could think of it as an ordinal variable and using an Ordinal Logistic Regression or Cumulative Link Model. These models would give odds ratios / proportional odds.
Examples:
- Ordinal Logistic Regression: https://stats.idre.ucla.edu/r/dae/ordinal-logistic-regression/
- Cumulative Link Model: http://cran.uni-muenster.de/web/packages/ordinal/vignettes/clm_article.pdf
I hope this helps already, but there might be better answer with more information about your study and data.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs