RMS Discussions

Along these lines for cv:

## Function to do r overall reps of B resamples, averaging to get
## estimates similar to as if r*B resamples were done

val <- function(fit, B, r) {
  z <- 0
  for(i in 1:r) z <- z + validate(fit, method='cross', B=B)[
          c("Dxy","Intercept","Slope","D","U","Q"),'index.corrected']
  z/r
}
val(fit, 10, 100)  # 100 repeats of 10-fold