Date:         Mon, 8 Sep 2003 08:38:25 -0700
Reply-To:     Stephen Ban <ban@zoology.ubc.ca>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Stephen Ban <ban@zoology.ubc.ca>
Subject:      Re: Anomaly in logistic regression output
In-Reply-To:  <sf5c9445.072@wpmail.paisley.ac.uk>
Content-Type: text/plain; charset="us-ascii"; format=flowed

Actually, I'm using a forward-step likelihood ratio.

Although I haven't run an iteration history, the number of steps between
the memory-sufficient and memory-deficient runs is about the same. Also,
since it is a forward-step procedure, the best explanatory variables should
be entering first (presumably, before memory usage becomes a problem), but
I think a difference is emerging even at these early steps.

Also, it seems curious that the procedure actually runs at all if SPSS is
truly running out of memory; previously, if I tried a procedure without
specifying the /EXTERNAL argument, SPSS would just terminate with an out of
memory error, which is not the case here, where it seems that SPSS is able
to complete the procedure by swapping to disk instead of RAM.

Stephen

At 02:37 PM 08/09/2003 +0100, Muir Houston wrote:
>I noticed that you used the backward Wald selection method in your
>regression - can you provide insght as to why this enter method was chosen?
>This might also impact on the number of iterations that Mark refers to.
>Just a thought and again curious for justification for enter method
>
>cheers
>muir houston
>
>
>
>Legal disclaimer
>--------------------------
>
>The information transmitted is the property of the University of Paisley
>and is intended only for the person or entity
>to which it is addressed and may contain confidential and/or privileged
>material.  Statements and opinions expressed in this
>e-mail may not represent those of the company.  Any review,
>retransmission, dissemination and other use of, or taking
>of any action in reliance upon, this information by persons or entities
>other than the intended recipient is prohibited.
>If you received this in error, please contact the sender immediately and
>delete the material from any computer.
>
>--------------------------