On 17 Mar 2000 06:16:53 -0800, [EMAIL PROTECTED] (Derek Ogle) wrote:
> Members,
>
> I could use help on two questions that I haven't been able to adequately
> answer for one of my students:
>
> 1) When performing a multiple linear regression we have examined t-tests for
> the null hypothesis that a coefficient is zero. If we examined the tests
> for each variable in the model, should you correct for multiple comparisons
> (e.g., use a Bonferroni-corrected alpha)?
The usual logic of having the variables in one regression is that they
make up a model -- so the overall F on the regression is the test on
the model, which is defined by *all* the variables. But if they are
effectively separate hypotheses, then, YES, you have "multiple
comparisons." If you can't form combinations of those variables, then
you might be stuck with something like Bonferroni corrections. Keep
in mind that the tail-area of the F-test in regression is increasingly
non-robust for the extreme. I don't think you take Bonferroni-style
beyond 10 or so variables.
>
> 2) When performing a multiple linear regression we have performed partial
> f-tests with the sequential SS (Type I SS) to examine if a particular
> variable "should be added" to a simpler model. If a series of these tests
> are used to find a parsimonious model that still fits should we correct for
> multiple comparisons?
"Stepwise inclusion" is usually a bad idea. See the comments in my
stats-FAQ, and their references. (If you are worried about correcting
for multiple tests, then you probably *shouldn't* add the variable
because it is probably capitallizing on chance.)
--
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/
===========================================================================
