Dear Hendrik,
the problem is that your var dep in the main equation does not have
missing values.
Like I don´t have better data, I will use this below with
transformations, oks. The important is that it works, not if make
sense!
***********begin example********
webuse sysdsn3, clear
preserve
rename age income
replace income = . if insure != 1
selmlog income nonwhite, select(insure=male nonwhite) showmlogit lee
selmlog income nonwhite, select(insure=male nonwhite) boot(50) lee
restore
*********end example****************
HTH,
Joao Lima
2008/11/28 Van Broekhuizen, H, Mnr <[email protected]> <[email protected]>:
> Hi
>
> I am currently writing my honours thesis and would like to execute the SEMLOG function is Stata.
> However, when I do Stata reports the message: "Beware: the mlogit step uses observations that have missing values in the main equation"
> Despite this message, Stata still proceeds to produce the mlogit output.
> However, Stata then reports the message: "Non-missing values of dependent variable for more than one outcome"
> And does not proceed to produce the output w.r.t the corrected main regression.
>
> I would appreciate it if anyone could clarify these error messages for me. What's wrong with my regression?
>
> Hendrik van Broekhuizen
> [email protected]
>
> *
> * 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/
>
--
----------------------------------------
Joao Ricardo Lima, D.Sc.
Professor
UFPB-CCA-DCFS
Fone: +553138923914
Skype: joao_ricardo_lima
----------------------------------------
*
* 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/
