Re: Linear regression and Logistic regression

On Wed, 08 Aug 2007 06:30:37 -0000, "akakiyakakievich@xxxxxxxxx"
<akakiyakakievich@xxxxxxxxx> wrote:


[...]

My questions in linear regression, about how can I compare two
distincts regressors to say one is "statistically" better than the
other? please correct me if I confuse. I'm using spss but my questions
are conceptual.
[snip]

I see to that spss has an option for select/discard variables in the
execution of the algorithm (forward, backward and stepwise selection),
are all greedy approachs (?), isn't necesarilly to exhaustive test all
variable combinations? (I know there is 2^n combinations, but for
example for n=5, 6, 7 may be useful) or isn't necesary test all to get
the optimal?

[snip]

If you can answer guide me refeering a website, document I'll be
gratefully.

Generally speaking, stepwise selection is a *very* bad idea for
most stat problems, unless you are in a position to do
massive replication and cross-validation.

See old comments from the stats-groups in my stats-FAQ.
Or use Google-groups to find relevant articles and threads --
try < groups:sci.stat.* stepwise author:ulrich > for a start.


--
Rich Ulrich, wpilib@xxxxxxxx
http://www.pitt.edu/~wpilib/index.html
.

Relevant Pages