Re: st: problems with browsing, dropping certain observations

[Ronan Conroy <rconroy@rcsi.ie>]

On 31 Beal 2010, at 12:58, Michael Findley wrote:

> I'm puzzled about what seems to be a simple formatting issue. I  
> import a .csv file with variables that range from .001 to .009. I  
> can carry out certain commands, such as generate a new variable that  
> has similar values, but I cannot perform other simple tasks such as  
> "browse if variable == .001" or "keep if variable == .001". The  
> variable is formatted as a float (%9.0g), which is a general format  
> that should allow this as I understand it. I was worried that  
> importing the data caused the problem, but when I just open stata  
> and enter similar values, I get the same thing. I also tried  
> reformatting the variable to several different fixed formats, but  
> still no luck. What am I missing? Any thoughts greatly appreciated.

Stata's internal representation of decimal numbers isn't exactly exact  
(the problem is mapping decimal numbers onto binary representations).

. set obs 9
obs was 0, now 9

. gen foo=0.001*_n

. list

     +------+
     |  foo |
     |------|
  1. | .001 |
  2. | .002 |
  3. | .003 |
  4. | .004 |
  5. | .005 |
     |------|
  6. | .006 |
  7. | .007 |
  8. | .008 |
  9. | .009 |
     +------+


. tab foo if foo==0.003
no observations

. tab foo if round(foo,0.001)==0.001

        foo |      Freq.     Percent        Cum.
------------+-----------------------------------
       .001 |          1      100.00      100.00
------------+-----------------------------------
      Total |          1      100.00


You can use the -round- function to make sure that Stata rounds the  
stored value to the exact number you want to match.


Ronan Conroy
=================================

rconroy@rcsi.ie
Royal College of Surgeons in Ireland
Epidemiology Department,
Beaux Lane House, Dublin 2, Ireland
+353 (0)1 402 2431
+353 (0)87 799 97 95
+353 (0)1 402 2764 (Fax - remember them?)
http://rcsi.academia.edu/RonanConroy

P    Before printing, think about the environment




*
*   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/