Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!hookup!swrinde!howland.reston.ans.net!news.sprintlink.net!redstone.interpath.net!sas!mozart.unx.sas.com!saswss
From: saswss@hotellng.unx.sas.com (Warren Sarle)
Subject: Re: Time based conditional probability... HELP!
Originator: saswss@hotellng.unx.sas.com
Sender: news@unx.sas.com (Noter of Newsworthy Events)
Message-ID: <D3LqB0.549@unx.sas.com>
Date: Mon, 6 Feb 1995 23:23:24 GMT
References:  <Pine.SOL.3.91.950204170320.20962E-100000@bingsun1>
Nntp-Posting-Host: hotellng.unx.sas.com
Organization: SAS Institute Inc.
Lines: 32


In article <Pine.SOL.3.91.950204170320.20962E-100000@bingsun1>,
Scott Hackett <br00372@bingsuns.cc.binghamton.edu> writes:
|> If someone could help point me in the right direction I would be forever
|> grateful to them.  I'm working on a project, trying to find confidence
|> in neural network output.  The first part was to work on finding the
|> conditional probability of an event happening, given some output from
|> a neural net.  The events were represented by vectors and were not time
|> based.  Now, however, my events are represented by a series of different
|> vectors over time.  I need to find the conditional probability of a time
|> based event happening, given some neural net output.  I just can't
|> find any way to do this.  I cannot use a Poisson distribution, because the
|> vectors in the series are different.  Is there any way to use a Poisson
|> distribution when the events over time are not the same?  Any references
|> or pointers would be GREATLY appreciated!

I am confused by the description of the problem, but I think Scott
Hackett is talking about the analysis of what statisticians call
point processes. See:

   Cox and Lewis (1966) _The Statistical Analysis of Series of
   Events_, London: Methuen.

   Karr (1986) _Point Processes and Their Statistical Inference_,
   NY: Marcel Dekker.


-- 

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.
