The help for -dstdize- tells me of a -saving(filename)-
option that saves the computed standard population distribution
as a Stata dataset that can be used in further analyses.
Nick
[email protected]
Peter Jepsen
>
> I have standardized crude incidence rates using 'dstdize', as
> described in
> the manual. A snippet of the output is here:
>
> Summary of Study Populations:
> year
> male N Crude Adj_Rate
> Confidence Interval
>
> --------------------------------------------------------------
> ------------
> 1977
> 0 2567052 9.739e-06 9.748e-06 [
> 5.875e-06, 1.362e-05]
> 1977
> 1 2512827 6.367e-06 7.836e-06 [
> 3.599e-06, 1.207e-05]
> 1978
> 0 2577115 6.208e-06 6.227e-06 [
> 3.137e-06, 9.318e-06]
> 1978
> 1 2519844 3.968e-06 5.532e-06 [
> 2.006e-06, 9.058e-06]
> 1979
> 0 2585949 4.254e-06 3.961e-06 [
> 1.600e-06, 6.322e-06]
> 1979
> 1 2525588 4.355e-06 5.122e-06 [
> 2.039e-06, 8.205e-06]
> .
> .
> . (continues to 2002)
>
> I would like to graph the gender-specific standardized
> incidence rates by
> year, but I find it very difficult to do because the adjusted
> rates are
> stored in a matrix, whereas the years and genders are not. I have no
> experience with matrices, but I managed to make the data
> useable through...
>
> matrix peter=r(se) \ r(ub) \ r(lb) \ r(adj) \ r(crude) \ r(Nobs)
> matrix peter=peter'
> svmat double peter, name(col)
>
> However, I had to add the years and genders to the correct
> rates using ...
>
> egen male=fill(0 1 0 1)
> egen year=fill(1977 1977 1978 1979)
>
> And that made it possible to graph all the things I wanted.
> But isn't there
> a better way to do this? I would love it if the 'Summary of Study
> populations' were stored in a dataset. I can't make sense of the
> 'dstdize.ado'-file.
*
* 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/
