I think Fred is correct. At first glance, the
problem lies in the code
if _by() & _bylastcall() & "`total'"!="" {
di _n in gr _dup(79) "_" _n "-> Total"
if "`level'"!="" {
local level "level(`level')"
}
if "`exposur'"!="" {
local exposur "exposure(`exposur')"
}
ci `varlist' if `alluse' `weight', /*
*/ `level' `poisson' `binomia' `exposur'
return clear
ret add
}
in which the lines
if "`exposur'"!="" {
local exposur "exposure(`exposur')"
}
have the effect (in his case) of converting the option
call internally to
exposure(ex(duration))
The error message doesn't really point in the right
direction here.
Fred Wolfe
> There appears to be a bug in -ci- when the by option AND the
> total option
> is used, as shown below.
>
> . ci firstcataract if lpuse == 1,exposure(duration)
>
> --
> Poisson Exact --
> Variable | Exposure Mean Std. Err.
> [95% Conf. Interval]
> -------------+------------------------------------------------
> ---------------
> firstcatar~t | 46209 .0077907 .0004106
> .0070066 .0086385
>
>
>
>
> -> sex = Female
>
> --
> Poisson Exact --
> Variable | Exposure Mean Std. Err.
> [95% Conf. Interval]
> -------------+------------------------------------------------
> ---------------
> cataract | 10369 .0058829 .0007532
> .0044999 .0075568
>
> -> sex = Male
>
> --
> Poisson Exact --
> Variable | Exposure Mean Std. Err.
> [95% Conf. Interval]
> -------------+------------------------------------------------
> ---------------
> cataract | 35840 .0083426 .0004825
> .0074237 .0093439
>
>
>
> . bys sex: ci cataract if lpuse == 1,exposure(duration) total
>
> -> sex = Female
>
> --
> Poisson Exact --
> Variable | Exposure Mean Std. Err.
> [95% Conf. Interval]
> -------------+------------------------------------------------
> ---------------
> cataract | 10369 .0058829 .0007532
> .0044999 .0075568
>
>
> -> sex = Male
>
> --
> Poisson Exact --
> Variable | Exposure Mean Std. Err.
> [95% Conf. Interval]
> -------------+------------------------------------------------
> ---------------
> cataract | 35840 .0083426 .0004825
> .0074237 .0093439
>
> ______________________________________________________________
> _________________
> -> Total
> time-series operators not allowed
> in option exposure()
> r(101);
*
* 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/
