| Re: Box-Whisker plots in IDL [message #55402 is a reply to message #55398] |
Mon, 20 August 2007 12:40   |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
teich@atmsci.msrc.sunysb.edu writes:
> Well, I am looking into the histogram procedure, but I am not getting
> what I think the 25th and 75th quartiles should be. It seems
> histogram is not so easy to master. What I am looking into is doing
> the following:
>
>
> data=randomu(sd,100)*100
> box plot needs min, max, median which are straight forward:
>
> min(data)
> max(data)
> median(data,/even)
>
> For the quartiles I am trying:
>
> lower_ind=where(data lt median(data,/even))
> upper_ind=where(data gt median(data,/even))
> qtr_25th=median(data[lower_ind(0):lower_ind(n_elements(lower _ind)-1)],/
> even)
> qtr_75th=median(data[upper_ind(0):upper_ind(n_elements(upper _ind)-1)],/
> even)
>
> However, I think this would work only for a monotonically increasing
> array. I am not sure how to get 'data' like that. If anyone wants to
> add to this, feel free.
I calculate it like this:
data=randomu(sd,100)*100
minVal = min(data)
maxVal = max(data)
medianVal = median(data,/even)
; Find the quartiles.
binsize = (maxVal - minVal) / 4.0
h = Histogram(data, BINSIZE=binsize, REVERSE_INDICES=ri)
qtr_25th = Median(data[ri[ri[0]:ri[2]-1]])
qtr_75th = Median(data[ri[ri[2]:ri[4]-1]])
Print, minVal, maxVal, medianVal, qtr_25th, qtr_75th
END
With 100 values I get this:
0.401314 98.0063 58.9402 20.0477 73.3419
With 10000 values, I get this, which leads me to think the
algorithm might be correct:
0.0249010 99.9960 49.9658 25.0268 74.8059
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
