R help - Nov 2007 - Extracting x-values from an ecdf

Dear R users,

Quick question on the use of ecdf for producing a cumulative  
distribution function.

After having produced a cumulative distribution using ecdf, I would  
like to extract the exact x-value for a given percentile.  For example  
I can ascertain the 1st and 3rd quartile using the summary function:

cawa.cdp <- ecdf(cawaocc$LEFF80)
plot(cawa.cdp, ylab="Fn(x)")
summary(cawa.cdp)

But, I would like write a function so that I can produce multiple  
ecdfs for >100 separate distributions and then extract the x-value (in  
this case LEFF80) for a given percentile (e.g., 80) for each separate  
distribution.  Then I could produce an average x-value for all the  
distributions for a given percentile.  I hope this is not too  
confusing.  Thank you for the help!
-- 
Benjamin Zuckerberg
Doctoral Candidate
State University of New York
College of Environmental Science and Forestry
Illick 244A, 1 Forestry Drive
Syracuse, New York 13210
Tele: (315) 470-6985
E-mail: bzuckerb at syr.edu

Benjamin Zuckerberg <bzuckerb <at> syr.edu> writes:
> After having produced a cumulative distribution using ecdf, I would  
> like to extract the exact x-value for a given percentile.  For example  
> I can ascertain the 1st and 3rd quartile using the summary function:
> 
> cawa.cdp <- ecdf(cawaocc$LEFF80)
> plot(cawa.cdp, ylab="Fn(x)")
> summary(cawa.cdp)
Well, if you tell me what type of Kava cawaocc$LEFF80 is? Please: Be poltite,
check the posting guide: examples should be self running.

set.seed(4711)
kn = knots(ecdf(rnorm(100)))
kn[50]


Dieter