Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: mysterious inaccuracy when adding big numbers


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: mysterious inaccuracy when adding big numbers
Date   Fri, 8 Apr 2011 16:01:37 +0100

No mystery. This is a precision problem as discussed at length about
once a month on this list.

Executive summary:

1. Read about it in sources indicated by -search precision-.

2. Use non-default data types, -double- or -long-, for big numbers.

3. Search the archives for problems similar to yours before posting.

Nick

On Fri, Apr 8, 2011 at 3:49 PM, Trang Nguyen <nqtrang.hanoi@gmail.com> wrote:
> Hi.
>
> I am working on a dataset with households as observations that are
> nested in communes, districts and provinces. I have variables
> - province: province number (3 digits)
> - district: district number within each province (max 2 digits)
> - commune: commune number within each district (max 2 digits)
> - household: household number within each commune (max 2 digits)
>
> I wanted to make a unique ID for each household that doesn't repeat
> across communes, districts and provinces, that also shows me all the
> province/district/commune information. So I did this:
> gen ID = province*1000000 + district*10000 + commune*100 + household
>
> I got a variable ID that is correct for the province, district and
> commune components, but the last two digits do not match the value of
> the household variable. Instead they are 04 or 12 or 20.
>
> Could someone please help me figure out why this is so? My output is
> below. Thanks much!
>
> . gen ID = province*1000000 + district*10000 + commune*100 + household
>
> . count if ID != province*1000000 + district*10000 + commune*100 + household
>  7932
>
> . format ID %15.0g
>
> . list province district commune household ID in 1/20
>
>     +------------------------------------------------------+
>     | province   district   commune   househ~d          ID |
>     |------------------------------------------------------|
>  1. |      101          1         3          1   101010304 |
>  2. |      101          1         3          2   101010304 |
>  3. |      101          1         3          4   101010304 |
>  4. |      101          1         3          5   101010304 |
>  5. |      101          1        17          4   101011704 |
>     |------------------------------------------------------|
>  6. |      101          1        17          5   101011704 |
>  7. |      101          1        17          6   101011704 |
>  8. |      101          1        17          8   101011712 |
>  9. |      101          1        17          9   101011712 |
>  10. |      101          1        17         10   101011712 |
>     |------------------------------------------------------|
>  11. |      101          1        17         11   101011712 |
>  12. |      101          3         3          3   101030304 |
>  13. |      101          3         3          4   101030304 |
>  14. |      101          3         3          5   101030304 |
>  15. |      101          3         3          6   101030304 |
>     |------------------------------------------------------|
>  16. |      101          3         3          7   101030304 |
>  17. |      101          5        11          3   101051104 |
>  18. |      101          5        11          6   101051104 |
>  19. |      101          5        11          9   101051112 |
>  20. |      101          5        11         10   101051112 |
>     +------------------------------------------------------+
>
> . codebook ID
>
> -----------------------------------------------------------------------------------
> ID                                                             (unlabeled)
> -----------------------------------------------------------------------------------
>
>                  type:  numeric (float)
>
>                 range:  [1.010e+08,8.231e+08]        units:  1
>         unique values:  1308                     missing .:  0/8341
>
>                  mean:   4.6e+08
>              std. dev:   2.6e+08
>
>           percentiles:        10%       25%       50%       75%       90%
>                           1.1e+08   2.1e+08   4.1e+08   7.1e+08   8.1e+08
>
> Thanks much!

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index