Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Graph from 'dstdize'-output


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Graph from 'dstdize'-output
Date   Sun, 31 Oct 2004 16:23:33 -0000

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 
n.j.cox@durham.ac.uk 

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/



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