Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!gatech!concert!sas!mozart.unx.sas.com!saswss
From: saswss@hotellng.unx.sas.com (Warren Sarle)
Subject: Re: Stats help
Originator: saswss@hotellng.unx.sas.com
Sender: news@unx.sas.com (Noter of Newsworthy Events)
Message-ID: <D1A4y5.19G@unx.sas.com>
Date: Fri, 23 Dec 1994 20:01:17 GMT
References:  <3d9k1b$p6s@pangea.ohionet.org>
Nntp-Posting-Host: hotellng.unx.sas.com
Organization: SAS Institute Inc.
Lines: 41


The appropriate newsgroup for this question would be sci.stat.consult.

In article <3d9k1b$p6s@pangea.ohionet.org>, kwhitele@sol.ashland.edu (Karl K. Whiteleather) writes:
|> I have data from a 2-year project on terrestrial salamanders.  I need to
|> get some direction on how to look at the following parameters and relations.
|>
|> Data collected was concerned with the...
|>
|> ...presence or absence...
|> (1 or 0)
|>
|> ...of three species of Ambystoma...
|> (A or B or C)
|>
|> ...in eleven pools...
|> (1 or 2 or 3 or... 11)
|>
|> ...with regard to concentration...
|> (0 ppm - n ppm)
|>
|> ...of twelve ion/ionic compounds.
|> (1 or 2 or 3 ... or 12)

If I understand this correctly, you have 11 cases (pools) with 3
binary dependent variables (presence of each species) and 12
independent variables (ion/ionic compounds). Since you have more
independent variables than cases, you cannot sort out the separate
effects of different compounds.

|> Can a neural net help, or is this more simple than it appears?

If you are interested only in predicting salamander presence in other
pools, you could possibly use a neural net. But what you really need
to do is state your research question more clearly.

-- 

Warren S. Sarle       SAS Institute Inc.   The opinions expressed here
saswss@unx.sas.com    SAS Campus Drive     are mine and not necessarily
(919) 677-8000        Cary, NC 27513, USA  those of SAS Institute.
