On Wed, 05 Apr 2000 14:14:29 GMT, Madewell <[EMAIL PROTECTED]>
wrote:
> The original question said "to compare distributions to sets of data"
> so I believe the discussion concerns the "Goodness of Fit" tests.
Oh, then the proper answer, at the start, by whoever responded then,
would have been to say,
Smirnov described statistics that are a "functions of the
vertical distance between two empirical distribution functions"
[Conover, 1971; pg. 195]. The KS test is not applicable, after all.
But what I said about the choice of alternative hypotheses is always
applicable.
>
> The power should be directly related to the sample size vs. test method
> used. I plan to run some tests to investigate this.
>
So, don't forget to state the range of alternatives you are testing.
If you are comparing tests of different scopes: What is interesting
is when
- you can beat the specific test (or come very close to it) with a
general test, given the specific alternative; or
- you can quantify, just how much better the specific test *does*, on
its specific alternative; or
- you can show that the specific test actually performs close to the
general test on a good range of "general" alternatives.
The last one has a subjective element, since different folks are apt
to have their own experiences that they would prefer to see weight
given to, of what the general alternatives ought to look like.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
