Next: 5.4Hypothesis Testing---I
Up: 5: Goodness of
Previous: 5.3Bartlett's Function---Overview
- From Equation (35) we have

- and using our earlier asymptotic result

- Therefore,
. Standard Deviation is
just
.
- In fact, for any number of parameters n,
-
has mean 0 and variance 1, and so the rule
(
for l Standard Deviations) is an appropriate
generalization of the rule that one increases
by
.
Geoffrey Fox, Northeast Parallel Architectures Center at Syracuse University, gcf@npac.syr.edu