Skip to content

Rank samples by breaks in hist and assign result as factor

2 messages · Frostygoat, David Winsemius

Frostygoat
# 
Hi R users,

My goal is to rank my samples according to how they fall out in a
histogram with 10 bins to produce a ranking for each sample according
to where it falls on the histogram, with a "1" to represent one tail
of the hist, a "10" to represent the other tail, and a "5" for the
median/mean. I have a number of different data sets to do this with
and in all cases I need this ranking between1-10 regardless of the
number of samples in each set.

e.g.,

x=rnorm(50)
h=hist(x,breaks=10)
h$breaks

[1] -2.0 -1.5 -1.0 -0.5  0.0  0.5  1.0  1.5  2.0

So, I want to add a factor to x where if x is less than -2.0, then new
factor="1"; if x is greater than -2.0 but x is less than -1.5, then
new factor="2", etc.

rank doesn't seem to help

Any tips?

Thanks much.
David Winsemius
#
On Dec 6, 2011, at 7:25 PM, Frostygoat wrote:

            

      
      
      
    

        
> xb <- as.numeric(cut(x, h$breaks))
 > xb
  [1] 9 2 6 9 9 9 6 5 4 7 8 6 6 9 4 8 5 4 9 7 5 9 5 5 4 2 6 7 4 6 4 9  
3 8 5 3 6 1 7
[40] 5 7 9 4 4 4 7 8 5 6 7

 > str(xdf)
'data.frame':	50 obs. of  2 variables:
  $ x    : num  1.577 -1.525 0.254 1.54 1.6 ...
  $ xbrks: num  9 2 6 9 9 9 6 5 4 7 ...

            

      
      
      
    

        

      
      
    

        
David Winsemius, MD
West Hartford, CT