Re: st: Bar graph with IQR bars

[Lars Folkestad <lfolkestad@health.sdu.dk>]

Dear Juan
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

Hope this helps.

lars  

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-.
>
>Dirk
>
>========================================
>Dr. Dirk Enzmann
>Institute of Criminal Sciences
>Dept. of Criminology
>Rothenbaumchaussee 33
>D-20148 Hamburg
>Germany
>
>phone: +49-(0)40-42838.7498 (office)
>        +49-(0)40-42838.4591 (Mrs Billon)
>fax:   +49-(0)40-42838.2344
>email: dirk.enzmann@uni-hamburg.de
>http://www2.jura.uni-hamburg.de/instkrim/kriminologie/Mitarbeiter/Enzmann/
>Enzmann.html
>========================================
>*
>*   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/