[Limdep Nlogit List] Grouped data models

[William Greene]

Dear Mikołaj: The limit values in the GROUPED DATA model may be
constants or variables. (See page E27-61 in the manual.) By this
device, you can specify individual specific limits for the model.
Instead of, say, ;LIMITS = 10000,20000,30000, you might say
;LIMITS=lower_i,Middle_i,high_i.
/Bill Greene

----- Original Message -----
From: "Mikołaj Czajkowski" <miq at wne.uw.edu.pl>
To: "Limdep and Nlogit Mailing List" <limdep at limdep.itls.usyd.edu.au>
Sent: Friday, September 19, 2008 5:45:39 AM GMT -05:00 Columbia
Subject: [Limdep Nlogit List] Grouped data models


Dear All,

I would like to apply GROUPED DATA for analysis of payment card
responses in a CV study. This is only one of the approaches to use, but
it seems reasonable, since dependent variable is completely censored.

The problem is that there were two different payment cards used in the
study, and so the intervals overlap.

I could not find any way to code individual-specific limits using
GROUPED DATA in Limdep. Perhaps some of you would have an idea how to
work around it?

I will be grateful for all ideas.

Best regards,


-- 
     Mikołaj Czajkowski

     Warsaw Ecological Economics Center
     Warsaw University
     http://www.woee.pl/

_______________________________________________
Limdep site list
Limdep at limdep.itls.usyd.edu.au
http://limdep.itls.usyd.edu.au