BBR Session 11: Nonparametric Tests Using the Proportional Odds Model

This is a place for questions, answers, and discussion about session 11 of the Biostatistics for Biomedical Research airing 2020-01-24 introducing the proportional odds model as a generalization of the Wilcoxon and Kruskal-Wallis tests. Session topics are listed here. The video will be in the YouTube BBRcourse Channel and will be directly available here after the broadcast. Power calculations are covered, and Bayesian ordinal logistic regression is briefly introduced.

1 Like

Is there a good way to use orm with no covariates?

Trying to use orm for this to obtain CI on the mean/median as shown in the course video, but without any covariates:

library(rms)
y <- round(runif(100,0,10))
dd <- datadist(y)
options(datadist = 'dd')
f <- orm(y~1)

Get the following error: Error in fitter(X, Y, offset = offs, penalty.matrix = penalty.matrix, :
object ‘kof’ not found

Sorry it doesn’t handle that. It should have. Not sure if lrm does either. The fit object does have the -2 log likelihood for the intercept-only model.

1 Like