R help - Oct 2003 - indexing a vector

Dear All:
I'd like to know how to sort and then index a vector of floats by several 
levels in R.
For example

>x<-rnorm(100)
> MyLevels<-quantile(x,probs=c(0,.5,1))
> MyLevels
         0%         50%        100% 
-2.11978442 -0.03770613  2.00186397

next i want to replace each x[i] in x by 1,2,3 or 4 depending on which 
quantile that x[i] falls. How do I do that in a "vector" fashion?

I tried something like
>factor(x,levels=as.numeric(MyLevels))
but that doesnt work.


Thank you very much,
Vlad

morozov wrote:
> Dear All:
> I'd like to know how to sort and then index a vector of floats by
several
> levels in R.
> For example
> 
> 
> 
>>x<-rnorm(100)
>>MyLevels<-quantile(x,probs=c(0,.5,1))
>>MyLevels
> 
>          0%         50%        100% 
> -2.11978442 -0.03770613  2.00186397
> 
> next i want to replace each x[i] in x by 1,2,3 or 4 depending on which 
> quantile that x[i] falls. How do I do that in a "vector" fashion?
I think rank() is the function you're looking for.

Cheers

Jason
-- 
Indigo Industrial Controls Ltd.
http://www.indigoindustrial.co.nz
64-21-343-545
jasont at indigoindustrial.co.nz

On Thu, 2003-10-02 at 12:40, morozov wrote:
> Dear All:
> I'd like to know how to sort and then index a vector of floats by
several
> levels in R.
> For example
> 
> 
> >x<-rnorm(100)
> > MyLevels<-quantile(x,probs=c(0,.5,1))
> > MyLevels
>          0%         50%        100% 
> -2.11978442 -0.03770613  2.00186397
> 
> next i want to replace each x[i] in x by 1,2,3 or 4 depending on which 
> quantile that x[i] falls. How do I do that in a "vector" fashion?
> 
> I tried something like
> 
> >factor(x,levels=as.numeric(MyLevels))
> 
> but that doesnt work.
> 
> 
> Thank you very much,
> Vlad

You might want to look at ?cut, which returns a factor based upon
defining breakpoints in a continuous vector. You can break the vector
'x' and relabel the levels to give you what you need.

For example:

x <- rnorm(100)

MyLevels <- factor(cut(x, quantile(x), include.lowest = TRUE), 
                   labels = 1:4)

table(MyLevels)
MyLevels
 1  2  3  4
25 25 25 25

HTH,

Marc Schwartz

morozov wrote:
>>x<-rnorm(100)
>>MyLevels<-quantile(x,probs=c(0,.5,1))
>>MyLevels
> 
>          0%         50%        100% 
> -2.11978442 -0.03770613  2.00186397
> 
> next i want to replace each x[i] in x by 1,2,3 or 4 depending on which 
> quantile that x[i] falls. How do I do that in a "vector" fashion?
Take a look at the "cut" function.


-- 
Ross Ihaka                         Email:  ihaka at stat.auckland.ac.nz
Department of Statistics           Phone:  (64-9) 373-7599 x 85054
University of Auckland             Fax:    (64-9) 373-7018
Private Bag 92019, Auckland
New Zealand