Re: st: too many values for -tabulate twoway, summarize-, but not -tabulate twoway-

[László Sándor <sandorl@gmail.com>]

Oh, thanks, Eric!

Shame I missed this in -limits-.

That said, now I only wonder if these limits changed from Stata 11. In
that environment I think we still could use a hack (mentioned in
earlier postings), with more unique values than this.

But I am just rambling here uselessly, I'll check it myself instead.

Thanks again,

Laszlo

On Wed, Apr 4, 2012 at 12:54 PM, Eric Booth <eric.a.booth@gmail.com> wrote:
> <>
>
> When you introduce the summarize() option to include the mean, SD, and Freq in the -tabulate- table cells, you hit a similar issue with Stata limits as if you tried to -tabulate- 2 categorical variables with too many categories (see -help limits- for -tabulate- and -tabulate, summarize- to those limitations).
>
> 2 solutions you might consider: (1) subset the data you run in a single -tabulate, summarize- or (2) create your table using other commands (like -collapse- and -reshape- below).
>
>
> ******************!
> clear
> set obs 10000
> g x = int(runiform()*50)
> g by = int(runiform()*20)
> g w = int(runiform()*100)
> g y = runiform()
>
>
> **This works without an error:
> tab x by [aw=w]
>
> **While this gives me r(134):
>        **adds  Means, Standard Deviations, Frequencies to table cells
> **tab x by [aw=w] , sum(y) //doesnt work
>
> **1. subset your data:
> tab x by [aw=w] in 1/100, sum(y) mean //works
> tab x by [aw=w] in 101/200, sum(y) mean //or use [if]
>
> *2. or collapse & reshape:
> preserve
>  collapse (mean) y (sd) ysd=y [aw=w], by(x by)
>  reshape wide y ysd, i(x) j(by)
>        **outsheet, export, etc.
> restore
> ******************!
>
>
> - Eric
> __
> Eric A. Booth
> Public Policy Research Institute
> Texas A&M University
> ebooth@ppri.tamu.edu
> +979.845.6754
>
> On Apr 4, 2012, at 11:25 AM, László Sándor wrote:
>
>> Hi,
>>
>> I am still fiddling around with -tab, sum()- (using Stata MP 12.1 for
>> mac). I encountered some curious behavior that I could not track down.
>>
>> This works without an error:
>> tab x by [aw=w]
>>
>> While this gives me r(134):
>> tab x by [aw=w], sum(y)
>>
>> Why should there be a difference? Are these limits documented somewhere?
>>
>> FWIW, at the end I paste how I generated data for my particular test run.
>>
>> Thanks,
>>
>> Laszlo
>>
>> For reference, here is the help on this error:
>>
>> [P]     error . . . . . . . . . . . . . . . . . . . . . . . .  Return code 134
>>        too many values;
>>        1) You attempted to encode a string variable that takes on
>>        more than 65,536 unique values.  2) You attempted to tabulate
>>        a variable or pair of variables that take on too many values.
>>        If you specified two variables, try interchanging them.
>>        3) You issued a graph command using the by option.  The
>>        by-variable takes on too many different values to construct
>>        a readable chart.
>>
>>
>> set obs 1000
>> set seed 19821009
>> g y = rnormal()
>> g y2 = rnormal()
>> g by = y2 > 0.5
>> g w = runiform()
>> g x = runiform()
>> *
>> *   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/