Re: st: RE: Continuous vs discrete variables.

[Jean-Benoit Hardouin <jean-benoit.hardouin@orscentre.org>]

Salut Amadou,
A solution is to use -inspect- and to use r(N_unique) which contains the 
number of modalities (without "missing values").
For example :
unab varlist: *
local binary
local categorical
local numerous
foreach i of varlist `varlist' {
  qui inspect `i'
   if `r(N_unique)'==2 {
       local binary `binary' `i'
   }
   else if `r(N_unique)'<5 {/*or to an other value*/
       local categorical `categorical' `i'
   }
   else {
      local numerous `numerous' `i'
   }
}
di "`binary'"
di "`categorical'"
di "`numerous'"

Best,
Jean-Benoit

adiallo5@worldbank.org a �crit :


> Maarten wrote:
>
>
>                                                                                 
>                                                                                 
>                                                                                 
> > Amadou:                                                                       
> > If your discrete variable is consists of zeros and                            
> > ones, than you usually don't have to. If one                                  
> > variable distinguishes between more than two                                  
> > groups, than you can use -xi- prefix.                                         
>                                                                                 
> ------------------------------------------------------------------------------- 
>
>
> Dear Maarten,
>
> Thanks for answering. Actually, I am not in a regression or summarizing context.
> I have a huge dataset and would like, among other things, to know which
> variables
> are dummies, which are categorical (more than two but countable modalities),
> which
> are discrete type (1,2,3,...) and which are continuous. I want to place each
> group in
> local macros for future use.
> Cheers.
> Amadou.
>
>
> *
> *   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/
>
>
>  
*
*   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/