12,274 followers
@danielhmajor Good question - here is a recent one that we did (see methods section) https://t.co/smuATBxkD0 ... Adding to what is stated, I would also look at change in Cox-Snell R2 when adding prognostic factors to existing ones, and give a sample size j