st: Re: log binomial regression Stata command

["Joseph Coveney" <jcoveney@bigplanet.com>]

Thomas Weichle wrote:

The logit and logistic commands are alternatives for the glm command
with binomial family and logit link.  The poisson command is an
alternative for the glm command with poisson family and log link.  Is
there an alternative Stata command that will produce a log binomial
regression (for calculating relative risks or risk ratios) instead of
having to use the glm command with binomial family and log link?

--------------------------------------------------------------------------------

The help file for -glm- has a hyperlink that points to a list of logistic-type
estimation (regression) commands in Stata.  You can call it directly from the
command prompt by typing: help logistic estimation commands .

The list contains one other command that will give risk ratios: -binreg , rr-
I'm not sure what the difference is between -glm , family(binomial)- and 
-binreg , rr-.  Maybe one calls the other.

I'm also not sure what you're seeking in an alternative to 
-glm , family(binomial) link(log)-.  Is there some postestimation command that 
you're looking to do that -glm- doesn't support?

There is one other command that I'm aware of in Stata that computes risk 
ratios, but it's not an estimation command:  -cs- and -csi-.  You can find out
more by typing

help cs

or 

help epitab

at the command prompt in Stata.

Joseph Coveney


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