# RE: st: monthly means & CI

 From "Nick Cox" To Subject RE: st: monthly means & CI Date Wed, 14 Mar 2007 13:49:03 -0000

```Svend had some very good advice here.

Here's some more.

0. Whatever you do, it seems that you should look at

females / (males + females)

and also

females / (males + females + unknown)

1. Plot your means against month.

2. If there's no seasonality there is little point
to splitting by month.

3. If there is seasonality, the tacit assumption that
separate months are mutually independent is unlikely. One
way forward is a generalised linear model with
binomial link and time of year as predictor.
A trigonometric approach might help. See

SJ-6-4  st0116  . . . .  Speaking Stata: In praise of trigonometric predictors
Q4/06   SJ 6(4):561--579                                 (no commands)
discusses the use of sine and cosine as predictors in
modeling periodic time series and other kinds of periodic
responses

Nick
n.j.cox@durham.ac.uk

Svend Juul

> Gaby Serdan wrote:
>
> I have data on deaths. I need to calculate the mean &
> CI of females in proportion to all population. Im
> trying first to create a variable for each month then
> take the total number of female per month and then
> divide by total number of deaths per month.
>
> - and Clive Nicholas gave suggestions.
> ---------------------------------------------------------------
>
> I understand that you want to estimate the proportion
> of females among the persons who died each month. The
> data you provided are a bit surprising for the purpose,
> with one female, three males, and 21 with unknown sex.
> To create some more illustrative data, I:
>

*
*   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/
```