--- Stas Kolenikov <[email protected]> wrote:
> Your answer to (2) is not correct: the fixed effect
> is equivalent to
> introducing dummy variables only for the linear
> model. For anything
> non-linear, the fixed effect is conditioning on this
> (for the -clogit- > model, you are conditioning on >
the number of 0's and 1's in each
> center), which is some complicated algebra.
Thank you Stas. I though that they were equal if there
was only a single independent categorical variable in
the model.
Then the question remains, am I justified in using
-clogit- and not dummies in this situation?
Regards,
Ricardo.
I am pretty sure that they are asymptotically equivalent, provided that the
number of strata (indicator terms for the centers) is small compared to the
total sample size. Or, to be more precise, that the number of observations
within each stratum is "large." If not, then the unconditional logistic
regression with dummies is biased.
