Dear All,
This is quite a stimulating discussion. Thanks to all.
1) While writing my previous response, I just realized that -collapse-
is probably a better deal.
collapse (mean) varlist, by (ethnicity gender)
And indeed it is, except that I need to transpose it Excel (purely for
exposition purpose). Though I learned a great deal, particularly from
Toy's suggestions.
2) It is still intriguing as to why tabstat allows two levels in one
situation and not in the other situation.Though I am not sure if
anything needs to be defended. To me clearly depends on the problem at
hand.
3) I was not aware of the xml_tab. Seems to be an immensely powerful
stuff. Can we use it after table/tabstat?
Best,
Prabal
On 11/11/09, Sergiy Radyakin <[email protected]> wrote:
> In defence of XML_TAB/ADePT consider examples here (choose a country,
> click a report):
> http://go.worldbank.org/X0PYVDDXB0
>
> e.g.:
> http://siteresources.worldbank.org/EXTEDSTATS/Resources/3232763-1252439241095/6426999-1253031414057/BGD_2006.xls
>
> Regards, Sergiy Radyakin
>
>
>
> On Wed, Nov 11, 2009 at 11:47 AM, Roy Wada <[email protected]> wrote:
>> It's redundant to do both but usually not incompatible. It would be
>> like -by:- program making a call to another -by:- program. outreg2
>> does this internally. It probably could have been done with tabstat
>> but that would have fractured the table across the screen under
>> current setup.
>>
>> Roy
>>
>>> Two notions of "byability" are being mixed up in this thread: One is the
>>> -by
>>> :- prefix ([D], p. 21) that is allowed with most Stata commands. The
>>> other
>>> one is more like the -over()- option allowed with commands like -mean-.
>>>
>>> -tabstat- allows both, with the former one being able to handle a
>>> -varlist-
>>> with more than one member, as shown in my earlier example, and the latter
>>> (also called -by()- in -help tabstat-) being able to take only a single
>>> -varname-...
>> *
>> * 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/
>
*
* 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/
