 |
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: income classes as dependent variable
At 08:43 AM 12/11/2008, marcel spijkerman wrote:
Dear Stata users,
My data set contains data on income and a number of characteristics
(age, education etc.). I want to explain individual income by the
characteristics. Income is not known exactly, it is only known to
which income class an individual belongs to. I have reasons to
believe that income is not normaly distributed. Can I simply use an
ordered logit/probit model or should I use another type of model.
and if so, do there exist stata modules to estimate such model.
As Martin suggests, -intreg- may be worth checking into. Note,
however, that -intreg- assumes the DV is normally distributed. But
check the example the manual gives for -intreg-. It notes that you
can log income, and illustrates how you can choose between intreg and
oprobit. When it is legitimate to use it, I think intreg is greatly
preferable to oprobit, but my understanding is that intreg is very
sensitive to violations of its assumptions.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/