[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: ...stepwise regression...
On 15/04/2004, at 16:33, Conroy <email@example.com> wrote:
If you really want a hands-off approach and have some money to spare
then have a look at Knowledgeminer X. It is certainly not a Stata
replacement but appears to be a niche tool for your sort of problem.
on 13/04/2004 10:42, firstname.lastname@example.org at
I have read the FAQ file on Stata Site. I share their (Harrel &
If you have 500 predictor variables you are probably in a fishing
view on this point, but I have to set a interpretation model of a
variable and I have more than 500 indipendent variables. I have
more correlate variables but I need a procedure to select the most
If I don't apply the sw linear regression I don't know how select the
variables. I have to do.
expedition. If so, you must think of some useful strategies for
these to a list of candidates.
If you insist on sw, get hold of Velleman's discussion of the topic in
the manuals to Data Desk. He provides an approach which attempts to
maintain statistical sanity at each step (as Ronan Conroy suggests in
his own discussion) rather than handing the whole task over to a
* For searches and help try: