I thought I understood this problem,
and that it was a problem to do with
monthly dates.
Now it seems that these dates are daily dates, but
that you want a tabulation by months.
So my latest suggestion is
expand 2
bysort serial : replace bclmstdt = bclmeddt if _n == 2
gen mbclmstdt = month(bclmstdt)
tsset serial mbclmstdt
tsfill
tab serial mbclmstdt
That way you are assured exact months, although
the tabulation ignores day within month.
Note that you will get a display of 2 in the table
if a person starts and ends within the same month.
Nick
[email protected]
> -----Original Message-----
Joao Pedro W. de Azevedo
>
> Many thanks for your help.
> I have another question.
> The format of my date variable is elapsed date.
> As an approximation I'm using 30.5 as my average month,
> however I would like
> to be more precise.
> I would like to know if any one could suggest me any
> procedure to increase
> the acuracy of this process.
>
>
> . preserve
> . gen diff = 1+round((bclmeddt- bclmstdt)/30.5)
> . expand diff
> . bysort serial : replace bclmsdt = bclmsdt[_n-1] + 1 if _n > 1
> . tab serial bclmsdt
> . save expanded
> . restore
>
*
* 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/
