Date:         Tue, 23 Jul 2002 07:45:52 +0000
Reply-To:     Jollier Nord <jollier_n@hotmail.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Jollier Nord <jollier_n@hotmail.com>
Subject:      MODEL SELECTION
Content-Type: text/plain; format=flowed

Hi listers
I have just jointed SPSS list. At present I am attemping to develop an ARIMA
forecasting model hourly cacbon monoxide concentration with 5 predictors
like wind speed (X1), wind direction(X2), temperature (X3), vehicle density
(X4), radiation (X5).
The general relationship between concentration (CO) and explanatory
variables that affect it is :

LN(C)= B1 + B2*LN(X1)+ B3*LN(X2)+ B4*LN(X4)+ B5*LN(X5)

What should I do to select a best model from a candidate set of models ?
Thank you in advance

_________________________________________________________________
MSN Photos is the easiest way to share and print your photos:
http://photos.msn.com/support/worldwide.aspx