Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# Re: st: imprecise totals

 From William Buchanan To "statalist@hsphsun2.harvard.edu" Subject Re: st: imprecise totals Date Wed, 22 Aug 2012 05:41:06 -0700

```Hi Lena,

What exactly is it that you are hoping to be "more precise"?  Would you like the results extended to 16 decimal places, or do you mean that you want them truncated to whole number values only?  What is it that you feel is imprecise about your results?  Have you considered that you are trying to estimate totals with multiply imputed data?

I suggest reviewing the FAQ for information on how to ask a more precise question that is more likely to yield the answer(s) you are looking for.

HTH,
Billy

Sent from my iPhone

On Aug 22, 2012, at 4:35, Lena Sperling <lenalindbjergsperling@gmail.com> wrote:

> Dear all,
>
> I need precise results for the totals in stata 12, but even though I
> change the format, I keep getting the following result:
>
> Code: mi estimate, saving(egyemprate): svy linearized: total emp1 emp2
> emp3 WAP1 WAP2 WAP3, cformat(%10.0f)
>
> Result:
> Multiple-imputation estimates     Imputations     = 10
> Survey: Total estimation          Number of obs   = 54043
>
> Number of strata  =         2     Population size = 118094013
> Number of PSUs    =      1010
> Average RVI     = 0.0000
> Largest FMI     = 0.0000
> Complete DF     = 1008
> DF adjustment:   Small sample     DF:     min     = 1006.01
> avg     = 1006.01
> Within VCE type:   Linearized             max     = 1006.01
>
>
> Total   Std. Err.     [95% Conf. Interval]
>
> emp1    1.69e+07    1134994      1.47e+07 1.92e+07
> emp2    2.16e+07    2006111      1.77e+07 2.56e+07
> emp3    2.79e+07    1397313      2.52e+07 3.07e+07
> WAP1    3.01e+07    1925429      2.64e+07 3.39e+07
> WAP2    3.97e+07    3387931      3.30e+07 4.63e+07
> WAP3    4.83e+07    2279071      4.38e+07 5.27e+07
>
> Does anyone know how to increase the precision?
>
> Thanks!
> Lena
> *
> *   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/
```