Skip to content

Leave one out Cross validation (LOO)

5 messages · Alex Roy, milton ruser, Frank E Harrell Jr

#
Alex Roy wrote:
I don't think that LOO adequately penalizes for model uncertainty.  I 
recommend the bootstrap or 50 repeats of 10-fold cross-validation.  See 
for example the validate and calibrate functions in the R Design package.

Frank
#
Alex Roy wrote:
With resampling you should be able to estimate any parameter including 
sigma.  The Design package's validate.ols function can estimate sigma 
using the bootstrap or c-v, penalizing for backward stepdown variable 
selection, although I have found some counter-intuitive estimates of 
sigma using Efron's optimism bootstrap.

Frank