How to rescale "linear predictor" in drawing nomogram with "rms" package in R?

时间:2015-05-24 18:55:47

标签: r regression

I am trying to draw a nomogram from a logistic regression in R by using the rms package, but currently I have a problem: indeed, I can get the nomogram, but the "linear predictor" axis ranges from -2.5 to +3, and I'd like to know whether I can make it range from 0 to 1 (i.e. from 0% to 100% of predicted probability for Y to be = 1). I guess I have to set the argument "lp.at=...", but I am not able to do it by myself, so I hope someone can help! Thank you a lot for answering.

Leonardo Frazzoni, MD

Università degli Studi di Bologna

1 个答案:

答案 0 :(得分:3)

It sounds as if you want to suppress the linear predictor scale altogether. You can do that with nomogram(..., lp=FALSE). To add a scale with predicted probabilities use nomogram(..., fun=plogis, funlabel='Predicted Risk'). Your question confused the linear predictor with your notion of a "final" output of logistic regression, the predicted risk.