Re: st: Bar graph with IQR bars

[Nick Cox <njcoxstata@gmail.com>]

Juan's interest in plotting medians with IQRs does sound to me like
interest in box plots.

In addition to -graph box-, check out -dotplot- and -stripplot- (SSC).

Nick

On Wed, Dec 28, 2011 at 7:57 AM, Lars Folkestad
<lfolkestad@health.sdu.dk> wrote:

> There are a couple of things to consider.
> Firstly is the plot you want the best for your needs?
> A lot of people on this list (The gurus if you like) dislikes these bar
> plots with error bars:
> Here is why:
> http://biostat.mc.vanderbilt.edu/wiki/pub/Main/TatsukiKoyama/Poster3.pdf
>
> But if you insist on the bars than the solution is here:
> http://www.ats.ucla.edu/stat/stata/faq/barcap.htm
>
> Where it sais:
> generate hiwrite = meanwrite + invttail(n-1,0.025)*(sdwrite / sqrt(n))
> generate lowrite = meanwrite - invttail(n-1,0.025)*(sdwrite / sqrt(n))
>
>
> You should write
> . Generate 75IQRvar1 = the value you found using summarize , d
> . Generate 25IQRvar1 = the value you found using summarize , d

Den 27/12/11 23.23 skrev "Dirk Enzmann" <dirk.enzmann@uni-hamburg.de>:

>>Juan wrote:
>>http://www.hsph.harvard.edu/cgi-bin/lwgate/STATALIST/archives/statalist.11
>>12/date/article-877.html
>>
>>Perhaps using boxplots a better alternative, see -graph box-.
*
*   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/