Allin Cottrell schrieb:
On Wed, 3 Mar 2010, Henrique Andrade wrote:
> I would like to suggest some improvements to the "gretl plot
> controls":
>
I've now addressed some of these points in CVS.
> (1) On the "X-axis" tab I think it would be better if we could insert
> dates (e.g. 2010:3) instead of numbers in the minimum and
> maximum fields when the option "manual range" is selected.
>
This is for time-series plots? I think we're better with what we
have. If you want to adjust the date range for the plot you can
set the sample. The current x-axis adjustment lets you fine-tune
the amount of extra space to the left and right of the data (by
default gretl leaves a little "elbow room" but this might be to
everyone's taste). Setting the x-range by date would be too coarse
for that in most cases.
> (2) On the "Main" tab I think it would be nice if we could define
> different font size for the Title of plot.
>
I don't think this has a very favorable cost-benefit ratio. If
you're going for publication quality it's best to leave the plot
untitled and let LaTeX (or word processor) handle the caption.
> (3) Include the option "Reference grid" in the GUI (Gnuplot
> commands: set grid, set grid front, set grid back).
>
There's now a grid button in the plot editor. But if you want the
grid on top of the data, you'd have to do that manually,
> (4) Include the option "Insert shaded area" in the GUI.
>
We now have something along these lines, for annual, quarterly and
monthly time series plots. It's implemented via a plain text file
that defines start and stop dates for the shaded bars. CVS and the
snapshots contain a sample file that lets you add NBER recession
bars, and you can write your own file on the same pattern.
Allin.
_______________________________________________
Gretl-devel mailing list
Gretl-devel(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-devel
Hey Allin,
Thanks for the implementation of shaded areas - a nice feature.
But some bugs remain. For some variables an error occurs if I want to
add the shaded area, for some others the shaded area fills only a part
of the graph whereas for some others everything is fine. Try for example
the money.gdt file. For "tbr" I get "data error", "m" works
fine, "infl"
produces also a "data error", for "y" the shaded area is not complete
(shaded only for the range 7.2-8.8) and also "cpr" produced a "data
error".
Best,
Artur