RMS Describing, Resampling, Validating, and Simplifying the Model

data analysis modeling strategy
, ,
188

Hi @f2harrell, could you kindly help me with the following?

I’m using rms 8.1-0.

c <- calibrate(m, B = 500, group = data, kint = 10)

How are the gray confidence lines drawn in plot(c)?
The object c has the columns “Lower” and “Upper”. Are these values subtracted from and added to, respectively, the column “calibrated.corrected”?

What’s the level of smoothing applied? I’m having trouble replicating in ggplot the smoothed calibration curve produced by plot(c).

Thank you.

189

Type ?calibrate to see information about the new confidence intervals or go directly to Bootstrap Confidence Limits for Bootstrap Overfitting-Corrected Model Performance – Statistical Thinking .Lower and Upper stand on their own. Look at the plot.calibrate code to see any smoothing method it uses.

1 Like