DMRT BY SAS.8

2
PROGRAM: VERIFY IF THERE ARE SIGNIFICANT DIFFERENCES AMONG VERIFICATION RUN (2) MEANS AND OPTIMIZATION RUNS (1) Data tondat; input x y1; LABEL x1=OPT; x2=VERIFICATION; cards ; 1 3.401905 1 3.401905 1 3.401905 2 3.867403315 2 4.162976085 2 4.079287561 ; proc glm ; class x; model y1 =x; means x / duncan ; run ; quit ; OUTPUT: The SAS System 3 05:42 Sunday, February 4, 2013 The GLM Procedure Duncan's Multiple Range Test for y1 NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate. Alpha 0.05 Error Degrees of Freedom 4 Error Mean Square 0.011605 Number of Means 2 Critical Range .2442 Means with the same letter are not significantly different. Duncan Grouping Mean N x A 4.03656 3 2 B 3.40191 3 1 ANALYSIS:

description

A PROGRAM TO CALCULATE DUNCAN'S MULTIPLE RANGE TEST BY SAS 8.0

Transcript of DMRT BY SAS.8

Page 1: DMRT BY SAS.8

PROGRAM: VERIFY IF THERE ARE SIGNIFICANT DIFFERENCES AMONG VERIFICATION RUN (2) MEANS AND OPTIMIZATION RUNS (1)

Data tondat;input x y1;LABEL x1=OPT; x2=VERIFICATION;cards;

1 3.4019051 3.4019051 3.4019052 3.8674033152 4.1629760852 4.079287561;proc glm;class x;model y1 =x;means x /duncan;run;quit;

OUTPUT:

The SAS System 3 05:42 Sunday, February 4, 2013

The GLM Procedure

Duncan's Multiple Range Test for y1

NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.

Alpha 0.05 Error Degrees of Freedom 4 Error Mean Square 0.011605

Number of Means 2 Critical Range .2442

Means with the same letter are not significantly different.

Duncan Grouping Mean N x

A 4.03656 3 2

B 3.40191 3 1

ANALYSIS:

DMRT SHOWS THAT MEANS OF VERIFICATION RUNS (2) ARE SIGNIFICANTLY HIGHER THAN THE OPTIMIZATION RUNS (1)