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: cmp command; mvtobit


From   "Aurélie P. Harou" <[email protected]>
To   [email protected]
Subject   st: cmp command; mvtobit
Date   Thu, 20 Mar 2014 14:14:39 -0400

To Whom It May Concern:

I am trying to estimate a bivariate tobit model with left censoring in
one model and left- and right-censoring in the other. I believe the
stata command mvtobit only allows for left-censoring at 0. I have
tried using the cmp command, but I get "Warning: regressor matrix for
duration_post_04_yr equation appears ill-conditioned" and do not get
any results for the full model. I get this even when I reduce the
number of RHS variables so as to get a condition number of 700.
Furthermore, when I specify the cmp command to estimate the same model
as the mvtobit command (i.e., left censoring at 0), I still get the
same warning message and the full model is not estimated (whereas it is estimated using mvtobit).

Why/how is this happening? How can I get the cmp command to converge?

Thanks in advance for any help!
Best,
Aurelie

---------------------------
Aurélie P. Harou, PhD
Earth Institute, Post-doctoral Fellow
Agriculture and Food Security Center
Columbia University, Lamont Campus
61 Route 9W, Lamont Hall, 2F
Palisades, NY 10964
c: 954-614-8161
[email protected]



*
*   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