Re: st: running rcpoisson command

[Steve Samuels <sjsamuels@gmail.com>]

Fatma:

You've stated that the purpose of your study is to describe the
cross-sectional association of the number of living children with
mother's current age and other covariates. If so, you do *not* have a
censored response, and -rcpoisson- is the wrong approach. (It would be a
poor tool to deal with censoring even if you had retrospective
information and wanted to model age- and birth order- specific birth
rates.)

Other possible approaches include multinomial logistic regression,
variants of negative binomial regression, and mean linear regression. If
you have survey data, you would use the survey versions of these
commands.

One wouldn't expect the number of children to change much once a woman
is past child-bearing age. But, even in older women, period effects such
the introduction of effective contraception could translate to age
differences. Also adult child mortality would reduce the number of living
children of older women. I'd recommend fractional polynomial models in
age to pick up all these effects.


Steve


On Nov 11, 2012, at 10:05 AM, Fatma Romeh wrote:

Dear statalist,

I'm having a problem with running the rcpoisson command. I want to regress the number of living children per woman on some variables such as (her current age, her education, etc.). The women that I have in my sample are aged between 15-49. I think my dependent variable is right censored for the the women who are not reached the end of their fertility period. My question is how to write the rcpoisson command such that Stata treats the women who are below age 44 as right censored? I appreciate any help with my question, many thanks.

Best Regards,

Fatma Romeh


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/