Re: pshistogram GMT4.5.11: strange results

Dear Paul,

my problem is solved: The option -F works perfectly for me, if I understand it 
right:


>
−F

Center bin on each value. [Default is left edge].
>





>
−I

Inquire about min/max x and y after binning. No plotting is done. Append o 
to output an ASCII table of the resulting x,y data to stdout. 
Alternatively, append O to output all x,y bin data even when y == 0.




My new understanding:
* "-Io" and "-IO" report as x in every case the first margin of the bin 
(independent of "-F"). The second margin of the bin is found adding the value 
given by "-W".
* "-F" forces centered binning, but the output of "-Io" resp. "-IO" shows in 
the x column not the center of the bin but its first margin.

So only the question after the rounding effects of "-F" remains, but in my 
case it is more or less irrelevantly.

Best Regards
Sigward

On Monday 27 January 2014 23:45:52 you wrote:
> On Jan 27, 2014, at 7:20 AM, Sigward Funke <[email protected]> wrote:
> > Hi,
> > 
> > first my questions related to pshistogram:
> > * Could there some influence of rounding and margins of the bins?
> 
> Since your values land exactly on the bin borders things may round
> differently.  Clearly you want centered bins instead (-F).
> 
> > * Can I force 0.1 binning centered at 0.5, 0.6, 0,7 ... 1.5, 1.6, 1.7?
> 
> Yes -F
> 
> > * Or more general: Can I define one margin of the first (or last) bin?
> 
> You mean let the first bin cover everything to the left of it and the last
> everything to the right [or down/up to some specified limit?].  No, but
> you could add a feature request on gmt.soest.hawaii.edu
> 
> I think perhaps pshistogram is backwards when it reports 0.45 etc.  In my
> thinking, without -F we have bins that go from 0.5 to 0.6, 0.6 to 0.7 and
> so on, so the bin centers would be 0.55, 0.65, etc. But that is not what
> is reported.  With -F we should have bins from 0.55 to 0.65, 0.65 to 0.75
> etc, with bin centers at 0.6, 0.7 etc.  So perhaps the output coordinate
> is messed up.  I will check.
> 
> -p
> 
> > Background: Struggeling with pshistogram I created following script:
> > 
> > ------- start test_pshistogram.sh ----------
> > pshistogram -Io -W0.1 -V <<END
> > 1.7
> > 1.4
> > 0.9
> > 0.8
> > 0.8
> > 0.7
> > 0.7
> > 0.6
> > 0.5
> > 0.5
> > 0.5
> > 0.5
> > END
> > -------  end test_pshistogram.sh  ----------
> > 
> > Calling this script I get:
> > 
> > $
> > $ ./test_pshistogram.sh
> > pshistogram: 12 points read
> > pshistogram: WARNING: 12 multiple modes found
> > pshistogram: Extreme values of the data :	0.5	1.7
> > pshistogram: Locations: L2, L1, LMS; Scales: L2, L1, LMS	0.8	0.7	0.6
> > 0.381385035698	0.29652	0.14826
> > 
> > pshistogram: min/max values are :	0.5	1.7	0	5
> > 0.5	5
> > 0.6	2
> > 0.8	2
> > 0.9	1
> > 1.3	1
> > 1.6	1
> > $
> > 
> > The lines beginning with "pshistogram" fit my expections, but in the
> > table of the last lines I would expect:
> > 
> > 0.5 4
> > 0.6 1
> > 0.7 2
> > 0.8 2
> > 0.9 1
> > 1.4 1
> > 1.7 1
> > 
> > Playing with the Option -F the confusion increases. Adding "-F" I get
> > 
> > 0.45	4
> > 0.55	1
> > 0.65	2
> > 0.75	2
> > 0.85	1
> > 1.35	1
> > 1.65	1
> > 
> > Despite of the offset of 0.05 this result looks better, but it isn't what
> > I want. Should I add 0.05 to the first column?
> > 
> > Any hint is welcome!
> > 
> > Best Regards
> > Sigward Funke

To unsubscribe, send the message "signoff gmt-help" to [email protected]