Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down at the end of May, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
"Kaulisch, Marc" <kaulisch@forschungsinfo.de> |

To |
<statalist@hsphsun2.harvard.edu> |

Subject |
AW: st: Catplot (ssc) labelsize in scheme? Stata 11 |

Date |
Tue, 11 Oct 2011 10:42:54 +0200 |

Nick, Yes, I can reproduce the problem that the gsize label value (in a scheme) is ignored with graph hbar as well. I will contact Stata Tech Support.... Best regards, Marc -----Ursprüngliche Nachricht----- Von: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Nick Cox Gesendet: Dienstag, 11. Oktober 2011 09:59 An: statalist@hsphsun2.harvard.edu Betreff: Re: st: Catplot (ssc) labelsize in scheme? Stata 11 This is I think a question for StataCorp because, as I said earlier, -catplot- does nothing to the labels. Exactly how schemes work is for StataCorp, but the principle is that schemes set defaults, yet commands can override them. That applies to built-in commands too. Clearly you can 1. Set an option in -catplot-: sysuse auto catplot rep78 catplot rep78, var1opts(label(labsize(large))) 2. Clone -catplot- and set your own default. 3. Use -graph hbar- directly bysort rep78 : gen frequency = _N by rep78 : gen tag = _n == 1 graph hbar (asis) frequency if tag, over(rep78, label(labsize(large))) 4. Write your own alternative. Nick On Tue, Oct 11, 2011 at 8:43 AM, Kaulisch, Marc <kaulisch@forschungsinfo.de> wrote: > > I can change the value of gsize label from vsmall or huge, but everytime the labsize in the graph medsmall. My scheme is a modification of s2color. > > Changing labsize manually under var2opts works perfectly... > > My code uses the following options: > percent stack asyvars scheme l1title legend and ytitle I cannot see > why these would intervene into the scheme... > > Puzzled... Nick Cox > Find > > scheme-s2color.scheme > > and the line > > gsize label medsmall > > Nick > > On Mon, Oct 10, 2011 at 10:27 PM, Nick Cox <njcoxstata@gmail.com> wrote: >> You are correct in implying that -catplot- just lets -graph hbar | >> bar >> | dot- choose defaults for labelsize. >> >> Similarly, there is no machinery in -catplot- affecting bar width. In >> practice, it seems that tuning the gaps is the best way of changing >> the widths. >> >> e.g. >> >> sysuse auto >> catplot rep78 >> catplot rep78, var1opts(gap(0)) >> catplot rep78, var1opts(gap(2)) >> >> In general, -catplot- is mainly a wrapper that calculates frequencies >> or percents and then lets -graph- do almost all the work. >> >> I've not devoted much time to knowing my way around the scheme files. >> >> Nick >> >> >> On Mon, Oct 10, 2011 at 10:11 PM, Kaulisch, Marc >> <kaulisch@forschungsinfo.de> wrote: >>> Nick, >>> >>> A similar question to my previous one regarding ciplot (ssc). Everytime I call -catplot- (ssc) the labelsize is medsmall. My programming skills are zero but in the catplot.ado I cannot find a specification of labsize; so in contrast to my earlier question, I think I am able to control the labelsize in a scheme. But where? >>> >>> Another way around, would there be a possibility to align actual bar >>> height/width with labelsize (in catplot) >>> - Is barwidth working with catplot? >>> >>> Thanks! >>> >>> >>> Marc >>> >>> P.S.: @David Radwin: Thanks for pointing me to the "Visual Guide", but it is in the office and I am far away from it... >>> >>> >>> -----Ursprüngliche Nachricht----- >>> Von: owner-statalist@hsphsun2.harvard.edu >>> [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Nick >>> Cox >>> Gesendet: Samstag, 8. Oktober 2011 17:27 >>> An: statalist@hsphsun2.harvard.edu >>> Betreff: Re: st: ciplot (ssc) - change markersymbol >>> >>> This isn't anything to do with schemes, because, as you report, whatever your -scheme- choices, -ciplot- would override them. >>> >>> Welcome to the world of programming! That is, if you want -ciplot- to default to circles, you need to clone the program and set your own defaults. It's now your program, except in so far as it was mine. You should use a different name. >>> >>> Evidently, I preferred -dh- as default marker for myself, at least when -ciplot- was written. >>> >>> Nick >>> >>> On Sat, Oct 8, 2011 at 3:50 PM, Kaulisch, Marc <kaulisch@forschungsinfo.de> wrote: >>> >>>> I am really puzzled and cannot solve the problem myself. Drawing >>>> graph with ciplot (from ssc) I get a diamond as marker >>>> (independently of scheme used). As I do some graphs with the >>>> statsby-strategy, where I get a circle as marker, and some with >>>> ciplot, I would like to get a circle for both types of graphs. >>>> >>>> Where do I have to change the markersymbol in my scheme? >>>> I do not want to extend every -ciplot- command by msymbol(circle). >>>> >>>> >>>> Marc >>>> >>>> P.S.: Is there any good introduction into the customizing schemes? * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

- Prev by Date:
**st: Panel Poisson 2 stage** - Next by Date:
**Re: st: Merging datasets** - Previous by thread:
**Re: st: Catplot (ssc) labelsize in scheme? Stata 11** - Next by thread:
**st: Panel Poisson 2 stage** - Index(es):