If I have a histogram and a kernel estimator of a density and I say it is O
(n ^(-1/2) ) .
Exactly what does that mean in terms of choosing the number of bins to have
a good estimator of the density.
Does it mean that I need sqrt(n) as the number of Bins?
What about if it was O (n^(- 2/5) )?
--
Thanks,
Jim.
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|