I am working on a project where I need to fit an ordinal logistic regression model (using R).
There already are R functions for doing it, such as porl (MASS package). However, since my dataset is not particularly large (we have lab data, not epidemiological data), I would like to try a robust approach - such as bootstrap-based or other methods - and compare with my current results.
I know that there are functions in R that allows you to make such analysis for a binomial outcome (such as the glmRob function in the robust package), but I am having a hard time find a way to do the same for a three-level ordered outcome, such as mine.
Does anyone have any information on this, such as functions for doing such analysis, code suggestions or references for me to search?
Have a look at this post:
http://r.789695.n4.nabble.com/Robust-standard-errors-in-logistic-regression-td803215.html
-Noel
I suggest you use Ordinal Logistic Regression. There is a function named polr in the MASS package that perform these.
What about Bayesian methods? The simplest approach would be using the rstanarm-package, where you simply prefix your function call with "stan_", so instead of MASS::polr() you use rstanarm::stan_polr().
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Software
- R Statistical Package