Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
esther adler <email@example.com>
Fri, 9 Sep 2011 20:33:28 +0100 (BST)
I am trying to reproduce the results in "Estimation of multinomial logit models with unobserved heterogeneity using maximum simulated likelihood" by Peter Haan and Arne Uhlendorff (The Stata Journal (2006) Vol. 6, Number 2, pp. 229–245). (http://www.stata-journal.com/article.html?article=st0104)
Everything goes well until page 236 when I type
"ml init start 0.5 0.5 0.5, copy"
and get the error message:
"initial vector: matrix must be dimension 7
Could anyone tell me what is the problem? (beyond just saying that this is a conformability error)
Note that I know that I could use the mixlogit package, but I actually want to check the results from mixlogit due to an issue I have with the way the panel log-likelihood is reported in mixlogit (it does not correspond with the one calculated using the predicted probabilities from mixlpred and I don't understand why).
* For searches and help try: