Skip to content

bootstrapping

6 messages · aaron.foley at students.tamuk.edu, Stephan Kolassa, David Winsemius

#
Hi Aaron,

try the argument "statistic=mean". Then boot() will give you the mean 
turn angle in your actual data (which appears to be 6 degrees, judging 
from what you write), as well as the means of the bootstrapped data. 
Then you can get (nonparametric) bootstrap CIs by 
quantile(boot$t,probs=c(.025,.975)). As far as I can see, there is 
really no need to look at sd().

A more interesting question would be how to deal with the fact that 
-180=+180, there may be something to think about here...

HTH,
Stephan


aaron.foley at students.tamuk.edu schrieb:
3 days later
1 day later
#
On Jan 20, 2010, at 10:23 AM, <aaron.foley at students.tamuk.edu> <aaron.foley at students.tamuk.edu
> wrote:

            
?colMeans
David Winsemius, MD
Heritage Laboratories
West Hartford, CT