Hi!
I am working with a model where I am simulating a variable n times.
Subsequently, I want to find some distributional characteristic of the
simulations, for which I need the individual densities for the simulated n
values.
I have tried using both the "normden(.)"-command and the "kdensity" to find
the densities for the values. My distributional characteristic sums over
functions of the individual simulated values involving the individual
densities, and the problem is that the the characteristic simply increases
with an increasing number n of simulated values! In theory the distribution
should not change with an increasing number of observations, but only become
more precise. And thus should imply that every single of the simulated
values should be assigned a smaller density as the number of values
increases.
I think I have tracked the problem to the density-calculation, where the
used routines apparently do not take account of the number of values. Has
anybody worked with a similar setting, or know more about the calculation of
densities from empirical data?
Cheers!
Jesper
*
* 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/