Getting Started Example for PROC LATTICE
/****************************************************************/
/* S A S S A M P L E L I B R A R Y */
/* */
/* NAME: LATCOV */
/* TITLE: Getting Started Example for PROC LATTICE */
/* PRODUCT: STAT */
/* SYSTEM: ALL */
/* KEYS: LATTICE */
/* PROCS: LATTICE */
/* DATA: */
/* */
/* SUPPORT: RAB UPDATE: */
/* REF: COCHRAN, W.G. & COX, G.M.(1957), "EXPERIMENTAL */
/* DESIGNS", 2ND EDITION, NEW YORK: JOHN WILEY & SONS. */
/* MISC: */
/* */
/****************************************************************/
/****************************************************************/
/* This example shows the covariance option with PROC LATTICE. */
/* the data set specifies a 3x4 triple rectangular lattice */
/* design (three orthogonal replications). This dataset has */
/* two response variables (Yield and Weight). Specifying the */
/* covariance option gives an ANOVA table for each variable */
/* plus an analysis of covariance table for Yield and Weight. */
/****************************************************************/
/* This dataset is from Cochran & Cox p. 418. */
/* The variable Weight is added for this example. */
data TestCov;
input Group Block Treatmnt Yield Weight;
datalines;
1 1 1 16 2.20
1 1 2 9 1.84
1 1 3 4 2.18
1 2 4 0 2.05
1 2 5 3 0.85
1 2 6 11 1.86
1 3 7 16 0.73
1 3 8 23 1.60
1 3 9 15 1.76
1 4 10 7 1.19
1 4 11 11 1.20
1 4 12 12 1.15
2 1 4 5 2.26
2 1 7 14 1.07
2 1 10 6 1.45
2 2 1 17 2.12
2 2 8 19 2.03
2 2 11 8 1.63
2 3 2 10 1.81
2 3 5 6 1.16
2 3 12 9 1.11
2 4 3 11 1.76
2 4 6 20 2.16
2 4 9 17 1.80
3 1 6 15 1.71
3 1 8 20 1.57
3 1 12 10 1.13
3 2 2 15 1.77
3 2 9 16 1.57
3 2 10 9 1.43
3 3 3 3 1.50
3 3 4 1 1.60
3 3 11 6 1.42
3 4 1 22 2.04
3 4 5 11 0.93
3 4 7 17 1.78
;
/* The COVARIANCE option is specified which produces an */
/* Analysis of Covariance table. */
proc lattice data=testcov cov;
var yield weight;
run;
/* The COVARIANCE option is not specified, so the */
/* Analysis of Covariance table is not produced. */
proc lattice data=testcov;
var yield weight;
run;
/****************************************************************/
/* These next steps provide another example of the covariance */
/* option w/ PROC LATTICE. This data set specifies a 5x5 square */
/* lattice design. There are two response variables (Yield and */
/* Height). Specifying the covariance option gives an ANOVA */
/* table for each variable plus an analysis of covariance table */
/* for Yield and Height. */
/****************************************************************/
data TestCov2;
input Treatmnt Yield Block Group Height;
datalines;
1 6 1 1 2.20
2 7 1 1 1.84
3 5 1 1 2.18
4 8 1 1 2.05
5 6 1 1 0.85
6 16 2 1 1.86
7 12 2 1 0.73
8 12 2 1 1.60
9 13 2 1 1.76
10 8 2 1 1.19
11 17 3 1 1.20
12 7 3 1 1.15
13 7 3 1 2.26
14 9 3 1 1.07
15 14 3 1 1.45
16 18 4 1 2.12
17 16 4 1 2.03
18 13 4 1 1.63
19 13 4 1 1.81
20 14 4 1 1.16
21 14 5 1 1.11
22 15 5 1 1.76
23 11 5 1 2.16
24 14 5 1 1.80
25 14 5 1 1.71
1 24 1 2 1.57
6 13 1 2 1.13
11 24 1 2 1.77
16 11 1 2 1.57
21 8 1 2 1.43
2 21 2 2 1.50
7 11 2 2 1.60
12 14 2 2 1.42
17 11 2 2 2.04
22 23 2 2 0.93
3 16 3 2 1.78
8 4 3 2 2.25
13 12 3 2 2.84
18 12 3 2 2.58
23 12 3 2 2.05
4 17 4 2 0.85
9 10 4 2 1.96
14 30 4 2 0.73
19 9 4 2 1.60
24 23 4 2 1.36
5 15 5 2 1.19
10 15 5 2 1.20
15 22 5 2 1.75
20 16 5 2 2.26
25 19 5 2 1.37
;
/* PROC LATTICE without the COV option */
proc lattice data=testcov2;
var yield height;
run;
/* PROC LATTICE with the COV option */
proc lattice data = testcov2 cov;
variables yield height;
run;