Skip to main content

New answers tagged

3 votes

Significantly different pseudo R2 values

The Nagelkerke pseudo $R^2$ is a simple function of the model likelihood ratio $\chi^2$ statistic, so to do formal tests of a subset of the variables, omit those variables and do a likelihood ratio $\...
Frank Harrell's user avatar
3 votes

Is it ok to restrict data to uncensored patient in order to compare models

As I understand the situation, you have two "deep learning" survival models whose performances you want to compare on the same new test set. You should use all of the data in the test set, ...
EdM's user avatar
  • 92.5k

Top 50 recent answers are included