Re: st: summarize, store multiple means?

[Joerg Luedicke <joerg.luedicke@gmail.com>]

If you want to create a matrix by hand, you could define a matrix and
then fill each cell one at a time by looping over variables and
variable values. Consider the following example where the means of 10
variables are put into 10 rows and 2 columns:

/**********************/
clear all
set obs 100 // getting some data
gen sex=runiform()>.5
forval i=1/10{
	gen var`i'=rnormal()
	}
	
mat t1=J(10,2,0) //Defining empty matrix

forval i=0/1{
local a=1
	foreach v of varlist var*{
		sum `v' if sex==`i', meanonly
		mat t1[`a',`i'+1]=r(mean) // filling each row, one at a time
		loc ++a
		}
	}
	
mat li t1
/**********************/

J.


On Wed, Jun 29, 2011 at 4:09 AM, D-Ta <altruist81@gmx.de> wrote:
> Good Morning,
>
> I would like to simply store multiple means as generated by the summarize
> command:
>
> sum var1 var2 .. var100
> matrix averages=r(mean)
>
> However, only the mean of the last (var100) is stored in that matrix. Is
> there anyway to store all the means or do I have to loop over all variables
> and store each mean value seperately in the matrix column?
>
> Ultimately, I would like to run a command like:
>
> bys sex: sum var1 var2 .. var100
>
> and store two set of means in two matrix columns, but I assume accessing
> r(mean) after that command would, again, only give me the very last mean
> entry.
>
> Cheers
>
> Darjusch
> *
> *   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/