To compare rologit and clogit in STATA, I fit the rologit model on the full ranking data for 9 alternatives. Secondly, the most preferred 5 alternatives were coded as 1; otherwise, 0 creating a new dependent variable and then fit clogit model. The estimates coefficients were stored as ‘Ranking’ and ‘Choice’ and fit the Hausman test (pp. 13, rologit-Rank-ordered logistic regression, stata.com). The results reported no coefficients in common; specify equations (matchlist) for problems with different equation names. r(498). I do appreciate your time and efforts to help in testing made it easy and solving the problem.

Can anybody of you share experiences and give references?