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 (June 2006, week 3)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Tue, 20 Jun 2006 09:17:03 +0200
Reply-To:     alte@uni-greifswald.de
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Dietrich Alte <alte@UNI-GREIFSWALD.DE>
Organization: Institut für Epidemiologie & Sozia
              lmedizin, EMA Uni Greifswald
Subject:      Re: Change Point Analysis
Comments: To: Tom Hulsey <hulseytc@MUSC.EDU>
In-Reply-To:  <200606191643.k5JFTDPf028670@mailgw.cc.uga.edu>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Try this:

* Simulate data---------------------------------------------------------------------------------; data sim; a1 = 0; b1 = 0.02; x0 = 250; * oder: x0 = (a1-a2)/(b2-b1); b2 = -0.02; a2 = a1 - x0*(b2-b1); put x0= a2=; do x = 1 to 500 by 0.5; error = normal(0) * 10 + 2; if x < x0 then y = a1 + b1 * x + error; else y = a2 + b2 * x + error; output; end; keep y x x0; run; proc sort data=sim; by x; run;

goptions reset=all; proc gplot data=sim; plot y * x; run; quit;

* cut time var into before/after break;; data sim; set sim; x1 = (x-x0) * (x LE x0); x2 = (x-x0) * (x GT x0); run;

* estimation; proc glm data=sim; model y = x1 x2; output out = pred predicted=y_hat LCL=LCL UCL=UCL LCLM=LCLM UCLM=UCLM; contrast 'beta x1 = beta x2' x1 1 x2 -1; * tests equality of estimates; run;quit;

* picture it; goptions reset=all; symbol1 v=dot h=0.2 c=black ; symbol2 i=join w=1 v=none c=red; symbol3 i=join w=1 v=none c=black l=2; symbol4 i=join w=1 v=none c=black l=2; proc gplot data=pred; plot (y y_hat LCLM UCLM) * x /overlay haxis=axis1 href=250; run; quit;

Tom Hulsey schrieb:

>Greetings: >Does anyone have code for change point analysis? I have two continuous >variables: vitamin D and calcium. They correlate positively, with >increases in D, calcium increases -- up to a point. Then even though D >continues to increase, calcium plateaus. I need to determine the change >point. Suggestions? Thanks. Tom Hulsey > > >

-- ---------------------------------------------------------------- DIETRICH ALTE, Dipl.-Statistiker, Dr. rer. med. Projektmanager "Study of Health in Pomerania (SHIP)" Institut für Epidemiologie & Sozialmedizin EMA-Universität Greifswald - Medizinische Fakultät Walther-Rathenau-Str. 48, D-17487 Greifswald, Germany URL www.medizin.uni-greifswald.de/epidem/ Phone ++49(0)3834-867713, Fax ++49(0)3834-866684 ----------------------------------------------------------------


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