Date: Fri, 19 Nov 2004 17:27:32 -0800
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: "David L. Cassell" <cassell.david@EPAMAIL.EPA.GOV>
Subject: Re: Applying Neural Nets in SAS EM for continous target variable
Content-type: text/plain; charset=US-ASCII
Sigurd Hermansen <HERMANS1@WESTAT.COM> sagely replied:
> I would not rule out using neural nets to predict really difficult
> in time series forecasting (such as turning points); nonetheless, the
> sources of variation in time series (trends, seasonality, and cycles)
> fit particularly well into a classification model. For example, serial
> correlation typically explains everything and nothing in a time series
Standard neural nets are generally equivalent to multivariate
techniques of one sort or another. Projection pursuit analysis, etc.
There has been some info on this in SAS-L in years past. So I wouldn't
recommend neural nets for a clear time series problem.
> The fact that SAS has a separate product for analyses of time series
> suggests that you should look at that first, and other time series
> packages such as RATS/CATS as well. Time series analysis may be the
> complex problem in statistics. I recall from way back that my
> instructor advised his students to delay looking at time series until
> a better grounding in cross section analyses. He did not mention
> as a possible short cut. I have not heard anyone else recommend neural
> either. The only common ground that I see for neural nets and some of
> frequency domain programs used to estimate cycles is that both fall
> general class of 'black box' modelling tools.
Also look at Autobox.
David Cassell, CSC
Senior computing specialist