Keith,
Follow the steps:
1. Find the analytical maximum in the standard manner (FOC), which will be a
function of the parameter of interest (in this case lambda)
2. Take the second derivative with respect to lambda (you should still have
lambdas at this point.
3. Evaluate the second derivative at the value derived in step 1 (i.e.
substitute for lambda the value you found in 1, so that you don't have any
lambdas in the equation).
4. Simplify, if possible.
5. Switch the sign of the result in 4.
I hope this is clear (and correct), since it's what I used and I think I derived
it correctly.
Oliver
I am still hung up on how to compute the fisher
information. Could
somebody briefly describe how it was computed in the example in
section? As I mentioned before, I have just been putting a negative
sign in front of the second derivative of the log likelihood
function, but this produces results that are not intuitive (such as -
n on the in-class example).
_______________________________________________
gov2001-l mailing list
gov2001-l at
lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
--
Graduate Student
Department of Government
Harvard University
Cambridge, MA