Re: st: nlogitrum and nlogit: unbalanced data

[John Fulton <fulton@ssc.wisc.edu>]

Thank you very much Richard.
I also noted the confusion with the "id" prefix variables in the code.
Not knowing what's really going on, I lacked the confidence to know 
which ones to rename.
John.



Richard Gates wrote:

> John,
>
> It looks like there is a bug in of -nlogitrum- when weights
> are given.  I am able to recreate the "weights" error that
> you experienced (sorry for not using weights when I tried this).
>
> Starting on line 102 of -nlogitrum- there is the following 
> code
>
> 	if "`weight'" != "" {
> 		tempvar chkwght 
> 		qui gen `chkwght' `exp' 
> 		cap bysort id: assert `chkwght' == `chkwght'[1] if `touse'
> 		if _rc {
> 			dis as err /*
> 		*/ "weights must be the same for all observations in a group"
> 			exit 407
> 		} 
> 	 	local wgt `"[`weight'`exp']"' 
> 	}
>
> replace id to `newgp' in the bysort  
> 	cap bysort `newgp': assert `chkwght' == `chkwght'[1] if `touse'
>
>
> Once I made this change the estimation completed. 
> If you have further problems, we can iron them out off of
> statalist.  
> -Rich
> rgates@stata.com
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>   
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/