thank you very much !
> >From this I take it that you want analogues of the
> standard deviation and the variance based on deviations
> >from the median, rather than the mean. Such measures can
> be defined, but they do not appear to be supported in Stata.
>
> -tabstat- can only tabulate statistics from the menu
> provided. For the kind of data you have it may be
> that a log transformation is appropriate and that
> in practice the geometric mean will be close to the median.
>
> Alternatively, with your own measures you might be
> best off <a
href="http://www.ntsearch.com/search.php?q=writing&v=55";>writing</a> -egen- functions and then using -tabdisp-
> for tabulation.
>
> Nick
> [email protected]
>
> Andreas Aschbacher
> >
> > I use :
> > tabstat dosis_corr,stats(mean range sd var sk k iqr)
> > /dosis_corr is dosis-value in radioactive measurement /
> > I suppose that sd is appropriate to arithmetic-mean
> > and var too - accordant sd = sqrt((1/(n-1))*sum[(xi - x_arithm.)^2]),
> > sum starts at 1 and goes to n/ x_arithmetic .... arithmetic-average .
> > my question:
> > how do I get: tabstat dosis_corr,stats(median range sd var sk k iqr)
> > where sd is a function of median and var too / sd(median)
> > var(median) //
>
> *
> * 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/
>
--
+++ NEU bei GMX und erstmalig in Deutschland: T�V-gepr�fter Virenschutz +++
100% Virenerkennung nach Wildlist. Infos: http://www.gmx.net/virenschutz
*
* 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/
