LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (August 2007, week 4)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Tue, 28 Aug 2007 20:11:27 -0700
Reply-To:     David L Cassell <davidlcassell@MSN.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         David L Cassell <davidlcassell@MSN.COM>
Subject:      Re: Proc GAM
In-Reply-To:  <200708242220.l7OEoNrB024472@malibu.cc.uga.edu>
Content-Type: text/plain; format=flowed

woods.steve@GMAIL.COM replied: >I was just trying to build a model with hospital data. As you know, day-of- >week has great impact on every kind of hospital data. I have nothing but >hospital data to build my model. however i am trying to insert as many as >external impacts that causes fluctuation on it. Peopel don't get sick or >do not go to hospital on Saturday's and Sunday's. Interestingly, it is so >low that no one would like to see the difference. Think about some one ask >you how many people will show up in this hospital based on the 5years of >history? > >That is my million dollar question?

Well, PROC GAM can give you a feel for how day-of-week drives your variable. But you may want to take the information gleaned from this and apply it in a simpler setting first. I like to try PROC REG / PROC MIXED level models first, until residual diagnostics tell me otherwise.

But don't you need to incorporate holiday information too, if you are seeing this kind of 'weekend' effect?

HTH, David -- David L. Cassell mathematical statistician Design Pathways 3115 NW Norwood Pl. Corvallis OR 97330

_________________________________________________________________ Tease your brain--play Clink! Win cool prizes! http://club.live.com/clink.aspx?icid=clink_hotmailtextlink2


Back to: Top of message | Previous page | Main SAS-L page