|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: orpobit vs ologit
Whatever the theoretical differences (in terms of assumptions about the
distribution of the latent response variable, conditional on covariates;
not the marginal distribution), in practice both will give you
substantively the same results. (The coefficients will be different
because they are normallized differently, but the size of effects in
terms of substantively meaningful things like predicted probabilities
will be indistinguishable.)
I vaguely recall seeing an analysis once that one would need something
like a million cases to be able to distinguish empirically whether one's
data fit one or the other model better . . .
--Nick Winter
n j cox wrote:
Really, you can't have looked very hard.
You need a good text. In Stataland, you might
as well start with the books by Scott Long,
with or without Jeremy Freese. See the Stata
bookstore for details. The manual entries
on these commands certainly cover the basics,
and so there is no point in recapitulating those
here.
What -ologit- and -oprobit- have in common
is a response variable that you consider
to be an ordered categorical variable. -rep78-
is a simple example in the auto data. It is
not measured, but categorical. It is ordered.
I don't know what data you have that are normally
distributed, but whatever the details that
is not relevant to either of these methods.
If your response variable is approximately
normal, it can't be an ordered categorical
variable. Also, I doubt that the marginal
distribution is particularly important for
choice of method.
Nick
[email protected]
Vanessa Mahlberg
could anybody please tell me the difference of ordered probit and
ordered logit regression modells. I couldn?t find a plausible
explanation of the difference of these two models.
My data is approximated normally distributed. So I think I should use
oprobit X Y? Am I right with this assumption?
*
* 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/
--
--------------------------------------------------------------
Nicholas Winter 434.924.6994 t
Assistant Professor 434.924.3359 f
Department of Politics [email protected] e
University of Virginia faculty.virginia.edu/nwinter w
PO Box 400787, 100 Cabell Hall
Charlottesville, VA 22904
*
* 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/
