This preview shows page 1. Sign up to view the full content.
Unformatted text preview: have to be included in the regression
model. There is also evidence of pure quadratic curvature.
Using Minitab:
Response Surface Regression: Replicates versus A, B
The following terms cannot be estimated, and were removed.
B*B The analysis was done using coded units.
Estimated Regression Coefficients for Replicates
Term
Constant
A
B
A*A
A*B Coef
0.10190
0.03862
0.02813
0.01463
0.01608 S = 0.00689372
RSq = 98.19% SE Coef
0.003083
0.001541
0.001541
0.003447
0.001541 T
33.053
25.054
18.249
4.244
10.432 P
0.000
0.000
0.000
0.000
0.000 PRESS = 0.00148511 RSq(pred) = 97.18% RSq(adj) = 97.83% Analysis of Variance for Replicates
Source
Regression
Linear
Square
Interaction
Residual Error
Pure Error
Total DF
4
2
1
1
20
20
24 Seq SS
0.051684
0.045656
0.000856
0.005171
0.000950
0.000950
0.052634 Adj SS
0.051684
0.045656
0.000856
0.005171
0.000950
0.000950 Adj MS
0.012921
0.022828
0.000856
0.005171
0.000048
0.000048 F
271.88
480.35
18.02
108.82 P
0.000
0.000
0.000
0.000 Unusual Observations for Replicates
Obs
6
16 StdOrder
6
16 Repli...
View
Full
Document
This document was uploaded on 03/20/2014 for the course EECS 6.780J at MIT.
 Spring '08
 DuaneBoning

Click to edit the document details