Re: st: level and distribution plot side by side

[Nick Cox <njcoxstata@gmail.com>]

I did some more work on "nice numbers" on logarithmic scales about a
year ago, but I need to go back and finish it.

A more systematic approach to a related (but not identical) problem is
discussed in

SJ-8-1  gr0032  . . . . . . .  Stata tip 59: Plotting on any transformed scale
        . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  N. J. Cox
        Q1/08   SJ 8(1):142--145                                 (no commands)
        tip on how to graph data on a transformed scale

http://www.stata-journal.com/sjpdf.html?articlenum=gr0032

and implemented in -mylabels- from SSC.

For your time scales, consider different formats for the time variable and also

SJ-7-4  gr0030  . Stata tip 55: Better axis labeling for time points/intervals
        . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  N. J. Cox
        Q4/07   SJ 7(4):590--592                                 (no commands)
        tip on labeling graphs for time points and time intervals

http://www.stata-journal.com/sjpdf.html?articlenum=gr0030

Nick

P.S. I didn't know (or had forgotten) about Rense Corten's blog. He
comments that Statalist "is so busy that it is close to worthless if
you are not looking for something specific".

On Sat, Jan 5, 2013 at 6:46 PM, annoporci <annoporci@gmail.com> wrote:

> 2) I discovered a very useful piece of code to have the labels of a log plot
> display powers of 10.
>
> It is on Rense Corten's blog (comments off):
> http://www.rensecorten.dds.nl/index.php/2011/03/labeling-logarithmic-axes-in-stata-graphs/
>
> It's similar to something you wrote on Statalist back in 2003:
> http://www.stata.com/statalist/archive/2003-11/msg00770.html
>
> I adapted it to make it a little more systematic, e.g. writing:
>
> forval x = `expmin'(`expstp')`expmax'{
>
> instead of say
>
> forval x = 2(1)5{
>
> and I have set up some calculations beforehand decide what the values of
> expmin, expstp, expmax should be, based on the data being graphed.
>
> The whole thing could be wrapped into a program (by someone more able than
> me).
>
> I hesitate to post my version of the code since the original was not written
> by me.
>
> There isn't an awful lot I can give back to Statalist now, but hopefully
> some day...
>
> Thanks Nick.
>
> --
> Patrick Toche.
>
>
>
>
>
>
> Here is an example.
>
> sysuse sp500, clear
>
> histogram close, horizontal xsc(reverse) normal freq ysc(off)
> fxsize(20) name(g1, replace) width(10)
>
> line close date, ysc(alt) yla(, ang(h)) name(g2, replace) xla(,
> format(%tddd_Mon)) xtitle("2001")
>
> graph combine g1 g2, ycommon imargin(small)
>
> However, this example does not imply that I just wrote down these
> commands. No; this is the cleaned-up version after making numerous
> small changes. Open a do-file editor with a script once you have
> commands that work, expect to have to make many tweaks. -name(,
> replace)- is a crucial detail.
>
> Nick
>
>
> On Sat, Jan 5, 2013 at 10:37 AM, Nick Cox <njcoxstata@gmail.com> wrote:
>>
>> See -help graph combine- and then the manual entry. In Stata terms,
>> you combine using graphs from -twoway histogram- and -twoway line-.
>>  Nick
>>  Patrick Toche <annoporci@gmail.com> wrote:
>>
>>> I saw this very nice graph: the main graph area has a twoway line of some
>>> very spiky data, in a standard horizontal orientation; immediately next
>>> to
>>> it, on the left-hand side, is a frequency plot displayed sideways.
>>>  Does anyone have sample code for this sort of display?
>>>  many thanks.
>>>  A picture speaks a thousand words:
>>> http://www5.picturepush.com/photo/a/11879063/640/11879063.jpg
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/