Date:         Thu, 22 Feb 2001 09:08:52 -0800
Reply-To:     Dale McLerran <dmclerra@MY-DEJA.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Dale McLerran <dmclerra@MY-DEJA.COM>
Subject:      Re: hierarchical regression
Comments: To: max8ca@yahoo.com
Content-Type: text/plain

Max,

With PROC REG you can use syntax like the following:

PROC REG;
  MODEL y = x1 x2;
  TEST x1=0, x2=0;
  MODEL y = x1-x4;
  TEST x3=0, x4=0;      /* test x3 & x4 with x1 & x2 in model */
  MODEL y = x1-x7;
  TEST x5=0, x6=0, x7=0;    /* test x5-x7 with x1-x4 in model */
RUN;

Similar tests can be constructed in PROC GLM using the CONTRAST
statement.  However, I think this is enough to get you going.

Dale


>Date:         Wed, 21 Feb 2001 09:12:41 -0800
>Reply-To: max8ca@YAHOO.COM
>From: max8ca@YAHOO.COM
>Subject:      hierarchical regression
>To: SAS-L@LISTSERV.UGA.EDU
>
>Hello,
>
>I need to run a regression in which three blocks of
>variables will be entered separately (i.e.
>hierarchical regression) and F change (after each
>additional block is entered) will be generated. How do
>I get F change using either GLM or REG, etc. Thanks.
>
>
>Max
>
>__________________________________________________
>Do You Yahoo!?
>Yahoo! Auctions - Buy the things you want at great prices! http://auctions.yahoo.com/




---------------------------------------
Dale McLerran
Fred Hutchinson Cancer Research Center
mailto: dmclerra@fhcrc.org
Ph:  (206) 667-2926
Fax: (206) 667-5977
---------------------------------------

------------------------------------------------------------
--== Sent via Deja.com ==--
http://www.deja.com/