Chiara Mussida <[email protected]>:
I think that means transitions from unemployed (U) to employed (E) or
not in the labor force (N). Offhand, that sounds like a case for a
competing risks survival analysis; see
http://home.fsw.vu.nl/m.buis/wp/survival.html
http://www.iser.essex.ac.uk/teaching/degree/stephenj/ec968/
help st
but maybe you only have one observation over time. In that case,
-mlogit- and others in that family may be of use.
On Fri, May 23, 2008 at 10:44 AM, Chiara Mussida <[email protected]> wrote:
>> On Fri, May 23, 2008 at 8:46 AM, Chiara Mussida <[email protected]> wrote:
>>> Dear All,
>>> I'm tryng to allow correlation between risks related to the labour market.
>>> I have 2 dependent binary variables for transitions from U to E, and
>>> from U to N, respectively. The covariates are the same for both the
>>> equations:
>>> is a bivariate probit the proper choice for my purpose (correlation
>>> between risks)? I have 2 dependent variables and the same RHS of the
>>> regression.
>>>
>>> Thanks a lot,
>>> chiara
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
