I have conducted a LCA on my CBC data. Looking at BIC and CAIC, my optimal amount of segments is 4. Using the posterior segment membership probabilities file, my segments are:
1. 25.9% (29 respondents)
2. 15.2% (17 respondents)
3. 27.7% (31 respondents)
4. 31.3% (35 respondents)
When I run a choice simulator, not all respondents are allocated to one of the four segments. About half of the respondents (54) are allocated to 'Other' rather than to their designated segment. Why is this and how do I fix this?
(This happened in both the Share Of Preference and Randomized First Choice simulator models.)
Thank you in advance for any help!