Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: micombine logit, odds ratio


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: micombine logit, odds ratio
Date   Wed, 3 Dec 2008 19:07:17 -0000

Richard is right that -micombine- is regarded as by its authors as
superseded by -mim-. 

Stuff remains on SSC or in any other Stata software archive for one or
more of at least the following reasons: 

(a) Because no one asked for it to be removed. 

(b) Because often newer stuff requires a later version of Stata than
users may have. That is, older stuff may remain useful to people who
cannot afford to upgrade. 

(c) Because people may wish to run other do files or programs referring
to it to produce identical or compatible results, or to avoid rewriting
code. 

(d) For comparison purposes. 

(e) It does no harm and takes up minimal space. 

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

Richard Williams

At 01:24 PM 12/3/2008, Xu, Haiyong wrote:

>I am using "ice" and "micombine logit" to run a logistic regression 
>from the imputed data. The "or" option in "logit" command doesn't 
>work. It outputs the coefficients instead of odds ratio. Could 
>someone help me out of this problem? My code is something like this.
>
>ice married male white grad_degree using impute, seed(1123) m(5)
>use impute, clear
>micombine logit married male white grad_degree, or

There is an eform option for micombine.  See the micombine help for
details.

You might also check out the -mim- command, available from SSC.  It 
looks to me like it is more powerful than micombine.  Both feature 
Royston as author or coauthor, so I am not sure why both continue to
exist.

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index