Newsgroups: comp.ai.neural-nets
From: Johan@bondsrus.demon.co.uk (Johan Christofferson)
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!hookup!swrinde!pipex!demon!bondsrus.demon.co.uk!Johan
Subject: Re: stock prediction : (
References: <1994Oct7.123142.10810@gdstech.grumman.com> <1994Oct19.222340.27526@hpcvusn.cv.hp.com>
Organization: CMA
Reply-To: Johan@bondsrus.demon.co.uk
X-Newsreader: Demon Internet Simple News v1.29
Lines: 10
Date: Thu, 27 Oct 1994 17:03:34 +0000
Message-ID: <783277414snz@bondsrus.demon.co.uk>
Sender: usenet@demon.co.uk

There has been quite a lot of good work (see Granger: Spectral
Analysis of Economic Time Series,  Princeton 1964) on showing 
that there is no cyclicality in price series.   Recently, Peters,
in Fractal Market Analysis, Wiley, 1994, seems to have evidence
of non-periodic cyclicality.

Forget the technical analysis texts. They are a waste of time.

-- 
Johan Christofferson
