Hint: The result of simulation is a (simulated) sampling distribution for
the quantity of interest (e.g. expected value, first difference, predicted
value).
If this doesn't help, take a look at the section notes (esp the r code) on
simulating theta and quantities of interest to see how we calculated
uncertainty.
Miya
On Sat, Mar 7, 2009 at 5:26 PM, Matt Clifford <matt_c at mit.edu> wrote:
Hi everyone,
I'm trying to understand the how to give standard errors/confidence
intervals as required in parts 2.3 and 2.4 of this week's problem set.
Is this simply a case of looking at the m-vector of estimates for the
expected difference/expected probabilities over m simulations and
using the sd() or quantile() functions on this vector to generate
standard errors or confidence intervals? Or am I missing something
more fundamental?
Many thanks,
Matt
_______________________________________________
gov2001-l mailing list
gov2001-l at
lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
--
Miya Woolfalk
Ph.D. Student
Harvard University
Government and Social Policy