Latent Class Model in STATA using lclogit__Error happened

[Gi-Young Chae]

hello? I am a novice to STATA.

After conjoint analysis, A problem occurred during Latent Class Analysis.

There are 6 conjoint properties (program, accessibility, accommodation type, facilities, privacy, price), and the levels are pro(3 levels), acc(3), stay(3), fac(3), privacy(2), price(3).

I used STATA 14.


ㅇ The command syntax is like this : lclogit choice pro pro1 acc stay stay1 fac fac1 privacy price, group(gid) id(pid) ncl(2) member(gender married household disease)

- Independent variable: choice
- Dependent variable: pro pro1 acc stay stay1 fac fac1 privacy price
- memberships : gender married household disease


ㅇ An error message appears like this: 3,000 (group size) take 1,000 (# positives) combinations results in numeric overflow; computations cannot proceed
r(1400);

error message description is like this

error. . . . . . . . . . . . . . . . . . . . . . . . return code 1400
numerical overflow;
You have attempted something that, in the midst of the
necessary calculations, has resulted in something too large
for Stata to deal with accurately. Most commonly, this is
an attempt to estimate a model (say with regress) with more
than 2,147,483,647 effective observations. This effective
number could be reached with far fewer observations if you
were running a frequency-weighted model.



I changed different memberships and did it again and again, I got the same results.
Please advise me from the seniors.
thank you

[Nick Cox]

Cross-posted and commented on at https://stackoverflow.com/questions/...error-happened

The error message is detailed. The model you have asked for enormously, mind-blowingly more complicated than is possible to fit. There is no work-around except starting with much, much simpler models.

Please note our cross-posting policy, which is that you should tell us about it. https://www.statalist.org/forums/help#crossposting

[Gi-Young Chae]

Cross-posted and commented on at https://stackoverflow.com/questions/...error-happened

The error message is detailed. The model you have asked for enormously, mind-blowingly more complicated than is possible to fit. There is no work-around except starting with much, much simpler models.

Please note our cross-posting policy, which is that you should tell us about it. https://www.statalist.org/forums/help#crossposting

I did cross-posted to the 'stackoverflow.com'.
overflow - Latent Class Model using lclogit__Error happened - Stack Overflow