st: aggregating data & local macros

["mac.stata" <mac.stata@gmail.com>]

Dear all,

As some of you have already noticed I am more of a beginner with Stata  
and therefore confront you most likely with pretty basic questions.  
Nevertheless, some have already helped me quite a bit (many thanks!)!  
But, unfortunately, there are two other questions right at my side:

1. aggregating data
I have records of individuals on the community level and these should  
be matched with data on the district level (I have district numbers I  
can match; I will use "joinby"). Which is the right way to go: a) have  
the two data sets put together with "joinby" and then try using  
"collapse" to aggregate the data on a district level (and then regress  
the data on the district level) or b) again–on the community level–put  
the data sets together first but then work with the Stata command  
"xi: ... ... ... i.districts"? As these are records of individuals  
containing socio-economic data it might get very difficult when using  
"collapse", I think. So, is the "xi:" actually the right way to go?  
Will this in any way cause troubles (from a scientific view of the  
approach)?

2. local macro
I wanted to create a local macro containing all variables to be used.  
So, instead of having << use A B C D E ... Z using "data.dta" >> I  
would like to have << use `local_var' using "data.dta" >>. The list of  
variables to use is long (about 30 to 40 variables, obviously less  
though than the ones I want to kick out, so no need for "drop"). Now,  
because it is a long list of variables I wanted to manually use line  
breaks. I therefore used the "#delimit ;"  and "#delimit cr" before  
and after the definition of the local macro << local local_var = " A B  
C D ... Z" >>. I thought this performs fine, also because I did not  
get any error but when I checked if all went right, it showed that  
after the third (of about five) line breaks the variables were  
ignored. So, Stata did recognize the local macro but it seems the  
local macro only contains about the first three lines of the variables  
I included. What am I doing wrong here? (I now use the "delimit"  
command within the standard "use" command which works fine).

Many thanks for your great help!

Andrea


*
*   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/