View previous topic :: View next topic |
Author |
Message |
Ashi
Joined: 30 Dec 2007 Posts: 28
|
Posted: Fri Oct 03, 2008 5:59 pm Post subject: misaligned error bars if uneven datasets |
|
|
When importing from excel, I had some cells which were blank due to having unknown error. E.g:
Y Error
1.1 0.4
1.2 0.6
1.7 0.3
1.8
1.9 0.4
3.4 0.5
When importing into dplot, the "blank" space is not imported. Now the error series has less datapoints.
When selecting the error series as the error, dplot does not check the x values, and instead simply matches the error bars by the index in the series, resulting in the error bars appearing on the wrong point.
(The 1.8 point gets the error bars from 1.9, 1.9 gets the error bars from 3.4, and the 3.4 value ends up with no error bars).
I think it would be better if the error bars were matched to data points by their x values, but if it's by index, then we should at least get a message saying "number of datapoints in series don't match" or something. Better yet, ideally a blank space in a dataset could be imported into dplot as a null value (right now I'm putting in error bars of "0" to get around the problem, but a null value would be preferable, with the idea being that no error bars would be shown for that point). |
|
Back to top |
|
|
DPlotAdmin Site Admin
Joined: 24 Jun 2003 Posts: 2310 Location: Vicksburg, Mississippi
|
Posted: Sat Oct 04, 2008 12:51 pm Post subject: |
|
|
Quote: | Better yet, ideally a blank space in a dataset could be imported into dplot as a null value (right now I'm putting in error bars of "0" to get around the problem, but a null value would be preferable, |
How big is "null"?
The mechanism for doing something similar is in there now (Options>Amplitude Limits), but is unfortunately ignored for error bar amplitudes. I'll change things such that if an error bar magnitude is outside those limits then no error bar is drawn. _________________ Visualize Your Data
support@dplot.com
Last edited by DPlotAdmin on Tue Nov 11, 2008 1:02 pm; edited 2 times in total |
|
Back to top |
|
|
Ashi
Joined: 30 Dec 2007 Posts: 28
|
Posted: Mon Oct 06, 2008 2:55 pm Post subject: |
|
|
Quote: | How big is "null"? |
The same as the y value divided by 0
It should show no error bars if it's null |
|
Back to top |
|
|
DPlotAdmin Site Admin
Joined: 24 Jun 2003 Posts: 2310 Location: Vicksburg, Mississippi
|
Posted: Mon Oct 06, 2008 3:52 pm Post subject: |
|
|
I think you probably understand this, but just for anybody else who stumbles across this thread:
When you import data into DPlot it is stored into an array of real numbers. There's no such animal as "null". You *can* define an amplitude that DPlot will ignore with Options>Amplitude Limits, but outside of that option there's no way to say "this value is meaningless". In the next release the Amplitude Limits values will be honored for error bar magnitudes. That means that in your Excel spreadsheet you'll need to enter some bogus value instead of a blank cell.
I understand in your case you'd prefer to have the error value tied to a specific X, and in the case of importing all of the data from Excel in one pass that makes sense. But in general that can't work: you can, for example, import the X,Y data (no error values) from Excel, then copy/paste error values (no associated X) from some other source. In that case the "error" curve would have X values of 0,1,2,3,... and have no relationship to the curve that you want to assign error values to. _________________ Visualize Your Data
support@dplot.com |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|