[Limdep Nlogit List] Extracting latent class membership - follow-up question

[Anna-Kaisa Kosenius]

Thank you.

How should I interpret the following membership probabilities that don't
make sense to me at all:
1) some of the respondents have zero probabilities for all three classes
2) the fifth respondent (see below) is in the class 1 and 3 with probability
close to one, how is that possible? Shouldn't the probabilities of the
respondent always sum up to one?

Here are the first 10 respondents' membership probabilities for 3 classes: 
0	0	0
0	0	0
0	0	0
0	0	0
0.999999	1.40021e-006	0.999997
2.82723e-006	0.00556675	0.994433
0.999989	1.55484e-006	9.32038e-006
2.96642e-005	0.0245403	0.97543
0.0625064	0.93283	0.00466341
0.999993	8.63623e-007	5.8132e-006


-Anna-Kaisa



Anna-Kaisa Kosenius
Doctoral student
Department of Economics and Management
P.O.Box 27
FIN-00014 University of Helsinki
Tel. +358 9 191 58511
anna-kaisa.kosenius at helsinki.fi