Hi all,

I am trying to resolve some MLE equations, but I am not sure if this is
possible.
Context: logistic regression with error in x.
Let say I have a variable x explaining YBIN a binary variable, through a
logistic model.
If I only observe X (Xi=xi+ui) instead of x. ui being a random measurement
error (unmeasurable)
I will get an attenuated set of estimates. (that's what I am studying).
I can see that the bias is stronger with stronger beta1 and of course
stronger measurement error.
I wonder if it's possible to get some sort of function relating the biased
betas to the true betas.
This would be possible in linear regression, but I wonder if this is true in
MLE.

Using the 2 MLE equations I get from each model (with x and with X), I can
relate the true betas to the biased betas, but only in sums. As I cannot
observe the xi's. I don not see easy way to resolve this.

Any idea?

JP
--




--





===========================================================================
This list is open to everyone.  Occasionally, less thoughtful
people send inappropriate messages.  Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.

For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================

Reply via email to