Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

st: Estimation selection model with selmlog/svyselmlog using sample weights


From   "David Wiesinger" <[email protected]>
To   [email protected]
Subject   st: Estimation selection model with selmlog/svyselmlog using sample weights
Date   Thu, 24 Nov 2011 18:47:14 +0100

Dear statlisters,

i'm apologising for being so imprecise, it's my first time that i'm posting in this forum.

First i was estimating my selection model using the original "selmlog" (written by Bourguignon et al) command and i got a result. Unfortunatly i wasn't able to include my probability weights.

For that reason, i was trying to use "svyselmlog" ( i got it form the homepage http://fmwww.bc.edu/RePEc/bocode/s and it was written by R. E. De Hoyos). 

with 
version 8: svyset [pweight=Gewicht]  
svyselmlog continuous choice, select( discrete choice), meth(dmf2) bootstrap(20) force

i got the error message

"PSU is equal to the number of observations, svybsamp2 is not necessary. Use -bsample-"

Unfortunatly i don't know, how to use -bsample-, becasue i couldn't find a hint in the help-file for svyselmlog.

Nevertheless i wanted to try out the procedure, so i created 2 random psu's and i rerun the model

version 8: svyset [pweight=Gewicht], psu(random_psu)
svyselmlog continuous choice, select( discrete choice), meth(dmf2) bootstrap(20) force

and i got an "conformability error" after some bootstrap steps. When the error occures, seems to be dependent on how many variables i put into the selection term and the used procedure. The less variables, the more steps can be calculated.
when i am estimating the first stage with svymlogit, the model converges and i get a result.
When i'm estimating the same model with selmlog, it works.

It is very likly, that i was making a stupid mistake, but for now i can't find it. How can i estimate the model with my weights only? 

thanks in advance,

David Wiesinger

-- 
NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurück-Garantie!		
Jetzt informieren: http://www.gmx.net/de/go/freephone
*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index