1                                The SAS System  11:34 Tuesday, January 20, 1998

NOTE: Copyright (c) 1989-1996 by SAS Institute Inc., Cary, NC, USA. 
NOTE: SAS (r) Proprietary Software Release 6.12  TS020
      Licensed to UNIVERSITY OF NORTH CAROLINA AT CHAPEL HILL, Site 0003944001.



1          TITLE1 "EXAMPLE3.SAS--Demonstrate simple LINMOD use";
2          *LET NOSOURCE2 = SOURCE2 ;
3          * Data are from Danford, Hughes, and McNee,              *;
4          * "On the Analysis of Repeated Measurement Experiments," *;
5          * Biometrics, 16, 1960, pp. 547-565.                     *;
6          
7          * 0. Define raw data file;
8          FILENAME IN01 "C:\PROGRAMS\LINMOD31\EXAMPLES\PAYNE.DAT";
9          DATA PAYNE;
10              INFILE IN01;
11              INPUT @7 GROUP $ INIT SCORE2 SCORE4 SCORE6 SCORE8 SCORE10;
12         * Create indicator variables ;
13         CONTROL = (GROUP="1");
14         LOW     = (GROUP="2");
15         MODERATE= (GROUP="3");
16         HIGH    = (GROUP="4");
17         

NOTE: The infile IN01 is:
      FILENAME=C:\PROGRAMS\LINMOD31\EXAMPLES\PAYNE.DAT,
      RECFM=V,LRECL=256

NOTE: 45 records were read from the infile IN01.
      The minimum record length was 31.
      The maximum record length was 31.
NOTE: The data set WORK.PAYNE has 45 observations and 11 variables.
NOTE: The DATA statement used 1.87 seconds.


18         PROC PRINT DATA=PAYNE;
19         
20         * 1. Define directory in which LINMOD source code has been stored;
21         %LET LMDIRECT = C:\PROGRAMS\LINMOD32\SOURCE\    ;
22         
23         * 2. Define SAS macro code needed;
24         %INCLUDE "&LMDIRECT.MACROLIB.MAC" / NOSOURCE2 ;
41         
42         * 3. Reduce raw data to a TYPE=CORR file named _CORRDS_
43              with characteristics required by LINMOD;
44         &PROCSSCP DATA=PAYNE ;

NOTE: The PROCEDURE PRINT printed page 1.
NOTE: The PROCEDURE PRINT used 0.61 seconds.

2                                The SAS System  11:34 Tuesday, January 20, 1998


45                   VAR CONTROL LOW MODERATE HIGH INIT
46                   SCORE2 SCORE4 SCORE6 SCORE8 SCORE10;
47         
48         * 4.1 Start IML with at least minimum space needed;
49         *     Use SHOW SPACE statement and larger WORKSIZE, SYMSIZE values
50               to tune performance with bigger problems;

NOTE: The data set WORK._CORRDS_ has 14 observations and 13 variables.
NOTE: The PROCEDURE CORR used 0.58 seconds.


51         PROC IML WORKSIZE=200 SYMSIZE=350;
Worksize    = 204800
Symbol size = 358400
IML Ready
52         
53         * 4.2 Initialize LINMOD IML code;
54         &LINMOD ;
NOTE: Module _DEFOPT_ defined.
NOTE: Module _EREXIT_ defined.
NOTE: Module _FITMOD_ defined.
NOTE: Module _GETCSS_ defined.
NOTE: Module _INITL_ defined.
NOTE: Module _MAKESS_ defined.
NOTE: Module _NMLIST_ defined.
NOTE: Module _PRPARM_ defined.
NOTE: Module _PVTYPE_ defined.
NOTE: Module _SCAN_ defined.
NOTE: Module _SETOPT_ defined.
NOTE: Module _TSTGLH_ defined.
NOTE: Module _UNIREP_ defined.
NOTE: Module _UNSCAN_ defined.
NOTE: Module _ZPIVOT_ defined.
NOTE: Module FITMODEL defined.
NOTE: Module GETCORSS defined.
NOTE: Module MAKESS defined.
NOTE: Module SETOPT defined.
NOTE: Module TESTGLH defined.
NOTE: Module NAMELIST defined.
NOTE: Module UTREND defined.
NOTE: Module UMEAN defined.
NOTE: Module UPOLY1 defined.
NOTE: Module UPOLY2 defined.
2197       
2198       * 5. Retrieve the file _CORRDS_ created in Step 3 ;
2199       RUN GETCORSS;
2200       
2201       * 6. Define the model and estimate primary parameters;
2202       INDVARS = { "CONTROL" "LOW" "MODERATE" "HIGH" "INIT" };
3                                The SAS System  11:34 Tuesday, January 20, 1998

2203       DEPVARS=NAMELIST("SCORE",2,10,2);
2204       *Last statement creates following matrix--
2205       *DEPVARS = { "SCORE2" "SCORE4" "SCORE6" "SCORE8" "SCORE10" };
2206       RUN FITMODEL;
2207       
2208       * 7.1 Conduct a test (and estimation) step;
2209       NOTE={"* MANOVA Test of Main Effect of Treatment,            *"
2210           , "* comparing each treatment group to the control group *" } ;
2211       PRINT ,NOTE ;
2212       C = { 1 -1  0  0   0 ,
2213             1  0 -1  0   0 ,
2214             1  0  0 -1   0 } ;
2215       THETARNM= { "C-LOW"  "C-MOD"  "C-HIGH" };
2216       *U defaults to Identity matrix, if NROW(U)=0;
2217       RUN TESTGLH;
2218       
2219       * 7.2 Conduct another test (and estimation) step;
2220       NOTE={"* Trends by Treatment Interaction,                    *"
2221           , "* comparing each treatment group to the control group *" } ;
2222       PRINT , NOTE ;
2223       *Re-use the same C matrix;
2224       *Note that all user inputs (C, U, THETA0, THETARNM, THETACNM, etc.)
2225        are never changed by LINMOD, and hence remain in existence;
2226       *May FREE THETACNM, for example, to avoid conflict
2227        between successive invocations of TSTGLH;
2228       TIMES ={2 4 6 8 10};
2229       RUN UPOLY1(TIMES  ,"Time",
2230                  USCORE , SCORENM );
2231       U = USCORE;
2232       THETACNM = SCORENM ;
2233       THETA0=J(NROW(C),NCOL(U),0);
2234       RESET PRINT;
2235       RUN TESTGLH;
Exiting IML.
NOTE: The PROCEDURE IML printed pages 2-33.
NOTE: The PROCEDURE IML used 3.52 seconds.


NOTE: SAS Institute Inc., SAS Campus Drive, Cary, NC USA 27513-2414