Bookmark and Share

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]

st: Convergence issue MLL


From   Lulu Zeng <[email protected]>
To   "[email protected]" <[email protected]>
Subject   st: Convergence issue MLL
Date   Mon, 3 Feb 2014 15:49:38 +1100

Dear Statalist,

I am having trouble converging in the maximum likelihood process.

I am trying to estimate a mixed logit model in Willingness to Pay
(wtp) space using the command gmnl written by Hole & Gu (2013).

I have 14 variables -- 12 of them are dummies; "bill" and "sq" being
exceptions. "bill" monetary variable in the range of 0-12, "sq" is a
constant.

My commands are as follows:

gen const = 1

constraint 1 [Mean]bill = -1  (so the gmnl model can be transformed
into the wtp specification)

constraint 2 [tau]_cons = 0  (fixed parameters as oppose to random)

matrix start = 1,2,6,2,2,3,2,2,2,2,2,3,2,0,0,0  (these values are
chosen based on the wtp obtained from the clogit in the preference
space)

matrix scale = 1,0,0,0,0,0,0,0,0,0,0,0,0,0   (dummies shouldn't be
scaled, so the only variable to be scaled is "bill")

gmnl y bill odo2 odo3 odo1 rwq2 rwq3 rwq1 bwq2 bwq3 bwq1 pin2 pin3
pin1 sq,group(gid) id(pid) het(const) constraint(1 2) from(start,
copy) nrep(500) scale (scale)

The log likelihood stopped changing after 10 iterations, and all the
further iterations give the same log likelihood value with the Stata
comment saying either "not concave" or "backed up").

It would be really appreciated if I could have your help on this,
thank you in advance.

Best Regards,
Lulu
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index