| Re: Last bin of a histogram ? [message #83885] |
Wed, 10 April 2013 11:20 |
John Correira
Messages: 25
Registered: August 2011
|
Junior Member |
|
|
On 04/10/2013 02:08 PM, David Fanning wrote:
> John Correira writes:
>
>> I thought that defined the start of the last bin. If you look at the
>> output locations, the last one is 99.5, which, again according to the
>> help, defines the starting location for each bin.
>
> Well, OK, that is the starting location of the last bin, then. But, it
> will be empty, since you told the function to ignore anything that might
> go into it.
>
> Cheers,
>
> David
>
I see.
So the following two calls will return the exact same set of bins, but
different histograms:
h1 = histogram(data,min=0.,max=99.5,binsize=.5,locations=l1)
h2 = histogram(data,min=0.,nbins=200,binsize=.5,locations=l2)
print, array_equal(l1,l2)
print, array_equal(h1,h2)
Brilliant.
|
|
|
|
| Re: Last bin of a histogram ? [message #83886 is a reply to message #83885] |
Wed, 10 April 2013 11:08  |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
John Correira writes:
> I thought that defined the start of the last bin. If you look at the
> output locations, the last one is 99.5, which, again according to the
> help, defines the starting location for each bin.
Well, OK, that is the starting location of the last bin, then. But, it
will be empty, since you told the function to ignore anything that might
go into it.
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thue. ("Perhaps thou speakest truth.")
|
|
|
|
| Re: Last bin of a histogram ? [message #83889 is a reply to message #83886] |
Wed, 10 April 2013 11:02 |
John Correira
Messages: 25
Registered: August 2011
|
Junior Member |
|
|
On 04/10/2013 01:50 PM, David Fanning wrote:
>
> It seems to me you are missing the fact that you set MAX=99.5 in the
> HISTOGRAM command. Or, am I missing something?
>
> Cheers,
>
> David
>
I thought that defined the start of the last bin. If you look at the
output locations, the last one is 99.5, which, again according to the
help, defines the starting location for each bin.
John
|
|
|
|
| Re: Last bin of a histogram ? [message #83890 is a reply to message #83889] |
Wed, 10 April 2013 10:50 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
John Correira writes:
> I'm trying to understand why HISTOGRAM selects the bins it does. The
> final part I don't understand is the last bin. Here is an example
>
> ; some fake data
> data = RANDOMU(systime(1),1000)*100
>
> ; make sure there are some values in the last bin
> while N_ELEMENTS(where(data ge 99.5 AND data lt 100.,count)) do $
> data = RANDOMU(systime(1),1000)*100
>
> ; how many points fall in the last bin?
> print, count
>
> h = histogram(data,min=9.5,max=99.5,binsize=.5,locations=locs)
> nbins = n_elements(h)
>
> ; starting locations of the last few bins
> print, locs[nbins-5:nbins-1]
>
> ; how many points in the last few bins
> print, h[nbins-5:nbins-1]
>
> Why is the last bin returned by the HISTOGRAM function zero? According
> to the help, the LOCATIONS keyword returns "the starting locations for
> each bin", so to my mind the last bin should include values that fall
> between locations[-1] and locations[-1]+binsize. I must be missing
> something obvious but I can't see it.
It seems to me you are missing the fact that you set MAX=99.5 in the
HISTOGRAM command. Or, am I missing something?
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thue. ("Perhaps thou speakest truth.")
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
