[Limdep Nlogit List] Best-worst estimation

[Data Analytics Corp.]

Bill,

Thanks for the advice on this.  I had to put the issue aside for a few 
weeks to address other clients, but now I'm back on it.  I do have 
NLOGIT (4.0) so I looked at the documentation.  It's confusing.  On page 
N10-9, it says at the bottom that ranks data may be used with RPLOGIT.  
But then on page N17-23, if says that ranks data is not available for 
it.  Which is it?  Am I overlooking something?

Assuming it can handle the problem, the coding would be a 1 for the best 
option, 3 for the worst option, and 2 for the unranked options.  There 
will actually be more alternatives that won't be shown in any one choice 
set.  In fact, there are 32 all together with only 4 shown at once in a 
choice set.  So the 28 unshown ones would be coded 0 - correct?

Thanks,

Walt

-- 
________________________

Walter R. Paczkowski, Ph.D.
Data Analytics Corp.
44 Hamilton Lane
Plainsboro, NJ 08536
________________________
(V) 609-936-8999
(F) 609-936-3733
dataanalytics at earthlink.net
www.dataanalyticscorp.com