It doesn't seem to be explicit
in documentation, but the -by()-
variable of -median- must be numeric.
If your id variable is string, use -encode-.
Nick
[email protected]
Andreas Aschbacher
> I am not sure about median - test with command (see) below:
> I have two columns:
>
> x - y- column
> / x-column contains ID-number /
> ID - numbers are :I090,M010,M013,M020,I010,M012
> / y-column is count-value /
> I090 94.97 1... observation
> M010 117.62 2.... observation
> M010 64.83
> M013 91.39
> M010 138.69
> M010 39.77
> M020 39.05
> M010 56.98
> I010 87.51
> M010 97.36
> M010 33.96
> M012 74.28
> .... there are much more observations with same ID-numbers
>
> I want to use following command:
> median varname [weight] [if exp] [in range] , by(groupvar) [ exact
> medianties(drop | above | below | split) ]
>
> in order to make median test in the following groups:
> I090,M010,M013,M020,I010,M012
> my Nullhypothesis should be: equal medians for all groups
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
