Download Analysis of a Split-Plot Experiment using SAS and PROC MIXED - Prof. Mervyn G. Marasinghe and more Study notes Statistics in PDF only on Docsity!
Example^ F18 SAS Programdata^ paper;do^ temp=^200 ,^225 ,^250 ,^275 ; do^ rep=^1 to^^3 ; do method=^1 to^^3 ; input^ strength @;output;end;end;end;datalines;30 34 29 28 31 31 31 35 3235 41 26 32 36 30 37 40 3437 38 33 40 42 32 41 39 3936 42 36 41 40 40 40 44 45; run ; proc^ mixed^ data=paper noclprint^ noinfo^ method=type
cl;;
class^ rep method temp;model^ strength =^ method^ temp methodtemp/ddfm=satterth;random^ rep repmethod;lsmeans^ method/diff^ cl^ adj=tukey;lsmeans^ method*temp/slice=method;contrast^ 'Temp:Linear Trend'^ temp -^3 -^1
contrast^ 'T1 vs T2 @ M1'^ temp^^1 -^1 0 0^ method*temp^^1 -
contrast^ 'T1 vs T3 @ M1'^ temp^^1 0 -^1 0^ method*temp^^1
0 -^1 0 0 0 0 0 0 0 0 0 ;
contrast^ 'T1 vs T4 @ M1'^ temp^^1 0 0 -^1^ method*temp^^1
0 0 -^1 0 0 0 0 0 0 0 0 ;
title^ 'Analysis of a Split-Plot Experiment using PROC MIXED'; run ;
SAS Output^ Analysis of a^ Split-Plot Experiment using PROC
MIXED^1 The Mixed ProcedureType 3 Analysis of VarianceSum of Source^ DF^ Squares^ Mean^ Square^
Expected^ Mean Square^ Error^ Term method^2 128.388889^ 64.^
Var(Residual) +^4 Var(repmethod)^ MS(repmethod)+^ Q(method,method*temp) temp^3 434.083333^ 144.^
Var(Residual) +^ Q(temp,methodtemp)^ MS(Residual) methodtemp^6 75.166667^ 12.^
Var(Residual) +^ Q(method*temp)^ MS(Residual) rep^2 77.555556^ 38.^
Var(Residual) +^4 Var(repmethod)^ MS(repmethod)+^12 Var(rep) rep*method^4 36.277778^ 9.^
Var(Residual) +^4 Var(rep*method)^ MS(Residual) Residual^18 71.500000^ 3.^
Var(Residual)^. Type 3 Analysis^ of^ VarianceErrorSource DF F^ Value^ Pr^ > F method 4 7.08^ 0.0485temp 18 36.43^ <.0001methodtemp 18 3.15^ 0.0271rep 4 4.28^ 0.1016repmethod 18 2.28^ 0.1003Residual..^.
ContrastsNum^ DenLabel DF^ DF^ F^ Value^ Pr^ > F Temp:Linear Trend 1 18 108.87^ <.0001T1 vs T2 @ M1 1 18 9.44^ 0.0066T1 vs T3 @ M1 1 18 35.29^ <.0001T1 vs T4 @ M1 1 18 32.90^ <.0001 Least Squares^ MeansStandardEffect method temp Estimate Error^ DF^ t Value^ Pr >^ |t|^ Alpha^ Lower^
- method^1 35.6667 1.2574 3. Upper
- 28.37 <.0001 0.05 32.1340 39.
- method^2 38.5000 1.2574 3.
- 30.62 <.0001 0.05 34.9673 42.
- method^3 33.9167 1.2574 3.
- 26.97 <.0001 0.05 30.3840 37.
- method*temp^1 200 29.6667 1.6044 9.
- 18.49 <.
- method*temp^1 225 34.6667 1.6044 9.
- 21.61 <.
- method*temp^1 250 39.3333 1.6044 9.
- 24.52 <.
- method*temp^1 275 39.0000 1.6044 9.
- 24.31 <.
- method*temp^2 200 33.3333 1.6044 9.
- 20.78 <.
- method*temp^2 225 39.0000 1.6044 9.
- 24.31 <.
- methodtemp^2 250 39.6667 1.6044 9. - 24.72 <. - methodtemp^2 275 42.0000 1.6044 9. - 26.18 <. - methodtemp^3 200 30.6667 1.6044 9. - 19.11 <. - methodtemp^3 225 30.0000 1.6044 9. - 18.70 <. - methodtemp^3 250 34.6667 1.6044 9. - 21.61 <. - methodtemp^3 275 40.3333 1.6044 9. - 25.14 <.
Analysis of a^ Split-Plot Experiment using PROC^ MIXED
(^3) The Mixed ProcedureDifferences of Least Squares MeansStandard Effect^ method^ temp^ _method^ _temp^ Estimate
Error^ DF^ t^ Value^ Pr^ > |t|^ Adjustment method^1 2 -2.
1.2295^4 -2.30^ 0.0825^ Tukey-Kramer method^1 3 1.
1.2295^4 1.42^ 0.2277^ Tukey-Kramer method^2 3 4.
1.2295^4 3.73^ 0.0203^ Tukey-KramerDifferences of Least Squares^ MeansAdj^ Adj Effect^ method^ temp^ _method^ _temp^ Adj P
Alpha^ Lower^ Upper^ Lower^ Upper method^1 2 0.
0.05^ -6.2469^ 0.5802^ -7.2151^ 1.
method^1 3 0.
0.05^ -1.6635^ 5.1635^ -2.6318^ 6.
method^2 3 0.
0.05^ 1.1698^ 7.9969^ 0.2015^ 8.9651 Tests of Effect^ SlicesNum DenEffect method DF DF^ F^ Value^ Pr >^ F methodtemp 1 3 18 15.50^ <.0001methodtemp 2 3 18 10.21^ 0.0004method*temp 3 3 18 17.03^ <.
