R help - Jul 2009 - 95% Confidence Intervals for AUC - $auc.samples from the Daim Package

Hi

I am trying to perform a bootstrap estimate of classification accuracy of a
logistic regression using the 'Daim' package in r using the code at the
bottom of this post, this all works great and I get the .632+ misclassification
accuracy, specificity, sensitivity, AUC etc etc but what I would like is to
access the list of AUC for each of the bootstrap samples as I need calculate the
95% confidence intervals for the AUC of the ROC curve for this data using this
model. I was hoping to get this from the 2.5% and 97.5% quartiles of the AUCs of
the 10,000 bootstrap samples that were calculated.

I have set returnSample to TRUE which I think should mean the data from each
bootstrap is saved and when I run auc(x) I get the following
> auc(x,auc.samples)
$auc.632p
[1] 0.8296788

$auc.632
[1] 0.8302014

$auc.loob
[1] 0.8216521

$auc.app
[1] 0.8455208

$auc.samples
list()

my question is how do I access the $auc.samples list() or write it to a csv file
or something similar to get the 2.5% and 96.5% quartile range from? or is there
another way to get the 95% CI for the AUC?

Many thanks in advance

Lara

Lara.harrup@bbsrc.ac.uk


############
library(Daim)

mylda <- function(formula,train,test){
model <- lda(formula,train)

predict(model,test)$posterior[,"pos"]
}

x<-Daim(OUTBREAK2 ~ COHESION.2 + COHESION.23 + ED.2 + PLAND.12 +
    PLAND.2 + PLAND.25 + PLAND.26 + ALTITUDE_MEAN + SLOPE_MEAN + LSI,
model=mylda,data=landscape,labpos=1,
control=Daim.control(method="boot",number=10000),returnSample=TRUE,
cutoff=0.71)

auc(x)
summary(x)


	[[alternative HTML version deleted]]