Warning: The Cubic model is aliased.

Fit Summary

Response 3: total fenolik

Mixture Component Coding is L_Pseudo.

Source Sequential p-value Lack of Fit p-value Adjusted R² Predicted R²

Linear 0,2463 0,2556 -1,1687
Quadratic 0,0990 0,9819 -0,2711 Suggested
Special Cubic
Cubic Aliased

Sequential Model Sum of Squares [Type I]

Source Sum of Squares df Mean Square F-value p-value

Mean vs Total 1,235E+06 1 1,235E+06
Linear vs Mean 4038,30 2 2019,15 2,03 0,2463
Quadratic vs Linear 3954,18 3 1318,06 54,64 0,0990 Suggested
Sp Cubic vs Quadratic 24,12 1 24,12
Cubic vs Sp Cubic 0,0000 0 Aliased
Residual 0,0000 0
Total 1,243E+06 7 1,775E+05

Select the highest order polynomial where the additional terms are significant and the model
is not aliased.

Model Summary Statistics

Source Std. Dev. R² Adjusted R² Predicted R² PRESS

Linear 31,54 0,5037 0,2556 -1,1687 17385,49
Quadratic 4,91 0,9970 0,9819 -0,2711 10189,69 Suggested
Special Cubic *
Cubic * Aliased

 Case(s) with leverage of 1.0000: PRESS statistic not defined.

Focus on the model maximizing the Adjusted R² and the Predicted R².
ANOVA for Linear model

Response 3: total fenolik

Source Sum of Squares df Mean Square F-value p-value

Model 4038,30 2 2019,15 2,03 0,2463 not significant
⁽¹⁾Linear Mixture 4038,30 2 2019,15 2,03 0,2463
Residual 3978,30 4 994,58
Cor Total 8016,60 6

⁽¹⁾ Inference for linear mixtures uses Type I sums of squares.

Mixture Component coding is L_Pseudo.

Sum of squares is Type III - Partial

The Model F-value of 2,03 implies the model is not significant relative to the noise. There is
a 24,63% chance that an F-value this large could occur due to noise.

P-values less than 0,0500 indicate model terms are significant. In this case A, B, C are
significant model terms. Values greater than 0.1000 indicate the model terms are not
significant. If there are many insignificant model terms (not counting those required to
support hierarchy), model reduction may improve your model.

Component Effects (Cox)

Gradient in Component Gradient Std Approx t for H₀ Prob >

Component
Reals Effect Error Gradient=0 |t|
A-teh putih 16,63 16,63 34,55 0,4813 0,6554
B-teh hijau 50,23 50,23 34,55 1,45 0,2196
C-teh hitam -66,86 -66,86 34,55 -1,94 0,1251

Component Effects (Piepel)

Gradient in Component Gradient Std Approx t for H₀ Prob > Gradient in

Component
Reals Effect Error Gradient=0 |t| Pseudo
A-teh putih 16,63 16,63 34,55 0,4813 0,6554 16,63
B-teh hijau 50,23 50,23 34,55 1,45 0,2196 50,23
C-teh hitam -66,86 -66,86 34,55 -1,94 0,1251 -66,86
Base Point

Component Real Point Actual Point Pseudo Point

A 0,3333 0,3333 0,3333
B 0,3333 0,3333 0,3333
C 0,3333 0,3333 0,3333

Fit Statistics

Std. Dev. 31,54 R² 0,5037

Mean 419,97 Adjusted R² 0,2556
C.V. % 7,51 Predicted R² -1,1687
Adeq Precision 3,7807

A negative Predicted R² implies that the overall mean may be a better predictor of your
response than the current model. In some cases, a higher order model may also predict better.

Adeq Precision measures the signal to noise ratio. A ratio of 3,78 indicates an inadequate
signal and you should not use this model to navigate the design space.

Model Comparison Statistics

PRESS 17385,49
-2 Log Likelihood 64,26
BIC 68,16
AICc 71,26

Coefficients in Terms of Coded Factors

Component Coefficient Estimate df Standard Error 95% CI Low 95% CI High VIF
A-teh putih 431,05 1 25,93 359,05 503,06 1,09
B-teh hijau 453,45 1 25,93 381,45 525,46 1,09
C-teh hitam 375,40 1 25,93 303,40 447,40 1,09

The coefficient estimate represents the expected change in response per unit change in factor
value when all remaining factors are held constant. The intercept in an orthogonal design is
the overall average response of all the runs. The coefficients are adjustments around that
average based on the factor settings. When the factors are orthogonal the VIFs are 1; VIFs
greater than 1 indicate multi-colinearity, the higher the VIF the more severe the correlation of
factors. As a rough rule, VIFs less than 10 are tolerable.
Final Equation in Terms of L_Pseudo Components

total fenolik =
+431,05 A
+453,45 B
+375,40 C

The equation in terms of coded factors can be used to make predictions about the response for
given levels of each factor. By default, the high levels of the mixture components are coded
as +1 and the low levels are coded as 0. The coded equation is useful for identifying the
relative impact of the factors by comparing the factor coefficients.

Final Equation in Terms of Real Components

total fenolik =
+431,05436 teh putih
+453,45436 teh hijau
+375,39897 teh hitam

Final Equation in Terms of Actual Components

total fenolik =
+431,05436 teh putih
+453,45436 teh hijau
+375,39897 teh hitam

The equation in terms of actual factors can be used to make predictions about the response for
given levels of each factor. Here, the levels should be specified in the original units for each
factor. This equation should not be used to determine the relative impact of each factor
because the coefficients are scaled to accommodate the units of each factor and the intercept
is not at the center of the design space.
 Normal Plot of Residuals

Design-Expert® Software
99
Trial Version
95
90

Normal % Probability
total fenolik 80
70

50

Color points by value of 30

20

total fenolik: 10
5

374,123 482,677
1

-4,00 -3,00 -2,00 -1,00 0,00 1,00 2,00

Externally Studentized Residuals

Box-Cox Plot for Power Transforms
Design-Expert® Software 9,2

Trial Version 9

total fenolik 8,8

Ln(ResidualSS)

8,6

Current transform:
None 8,4

8,2
Current Lambda = 1
Best Lambda = -3 8

-3 -2 -1 0 1 2 3

Right-click to export full text of legend Lambda

Predicted vs. Actual
Design-Expert® Software 500

Trial Version 480

460

total fenolik
440
Predicted

Color points by value of 420

total fenolik: 400

374,123 482,677
380

360

360 380 400 420 440 460 480 500

Actual
 Residuals vs. Predicted
Design-Expert® Software 8,00

6,57968
Trial Version 6,00

Externally Studentized Residuals

4,00

total fenolik 2,00

0,00 0

Color points by value of

-2,00
total fenolik:
374,123 482,677 -4,00

-6,00
-6,57968

-8,00

360 380 400 420 440 460

Predicted
Residuals vs. Run
Design-Expert® Software 8,00

6,57968
Trial Version Externally Studentized Residuals 6,00

4,00

total fenolik 2,00

0,00 0

Color points by value of

-2,00
total fenolik:
374,123 482,677 -4,00

-6,00
-6,57968

-8,00

1 2 3 4 5 6 7

Run Number
Residuals vs. A:teh putih
Design-Expert® Software 8,00

6,57968
Trial Version 6,00
Externally Studentized Residuals

4,00

total fenolik 2,00

0,00 0

Color points by value of

-2,00
total fenolik:
374,123 482,677 -4,00

-6,00
-6,57968

-8,00

0 0,25 0,5 0,75 1

A:teh putih
 Cook's Distance
Design-Expert® Software 2,5

Trial Version
2

total fenolik

Cook's Distance
1,5

Color points by value of 1 0,940534

total fenolik:
374,123 482,677 0,5

0 0

1 2 3 4 5 6 7

Run Number
Leverage vs. Run
Design-Expert® Software 1,00

Trial Version 0,857143

0,80

total fenolik
0,60
Leverage

Color points by value of 0,40

0,428571

total fenolik:
374,123 482,677 0,20

0,00 0

1 2 3 4 5 6 7

Run Number
 DFFITS vs. Run
Design-Expert® Software 2 1,96396

Trial Version 1

0 0

total fenolik
-1

DFFITS
Color points by value of -2
-1,96396

total fenolik: -3

374,123 482,677
-4

-5

1 2 3 4 5 6 7

Run Number
DFBETAS for A vs. Run
Design-Expert® Software 2

Trial Version 1
1,13389

0 0

total fenolik
DFBETAS for A

-1 -1,13389

Color points by value of -2

total fenolik: -3

374,123 482,677
-4

-5

1 2 3 4 5 6 7

Run Number

Report

Influence
Internally Externally
Run Actual Predicted Cook's on Fitted Standard
Residual Leverage Studentized Studentized
Order Value Value Distance Value Order
Residuals Residuals
DFFITS
1 443,91 419,97 23,94 0,143 0,820 0,778 0,037 0,318 7
2 374,12 375,40 -1,28 0,676 -0,071 -0,062 0,004 -0,089 3
3 482,68 442,25 40,42 0,276 1,507 1,984 0,289 1,225 4
4 394,37 414,43 -20,06 0,276 -0,748 -0,698 0,071 -0,431 6
5 409,88 403,23 6,65 0,276 0,248 0,216 0,008 0,134 5
6 435,29 453,45 -18,16 0,676 -1,012 -1,016 0,713 -1,468 2
7 399,54 431,05 -31,52 0,676 -1,756 -3,178 2,147⁽¹⁾ -4,593⁽¹⁾ 1

⁽¹⁾ Exceeds limits.

Design-Expert® Software A: teh putih
Trial Version
1
Component Coding: Actual

total fenolik
Design Points
374,123 482,677

X1 = A: teh putih
X2 = B: teh hijau
X3 = C: teh hitam
0 0

420

440

400

380

1 0 1
B: teh hijau C: teh hitam
total fenolik
Design-Expert® Software A: teh putih
Trial Version
1,000
1
Component Coding: Real

total fenolik
Design Points
374,123 482,677

X1 = A: teh putih
X2 = B: teh hijau
X3 = C: teh hitam
0,000
0 0
0,000

420

440

400

380

1,000
1 0,000
0 1,000
1
B: teh hijau C: teh hitam
total fenolik
Design-Expert® Software Predicted vs. Actual
Trial Version
500

total fenolik

480
Color points by value of
total fenolik:
374,123 482,677
460

440

Predicted
420

400

380

360

360 380 400 420 440 460 480 500

Actual
Design-Expert® Software
Trial Version
Component Coding: Actual

total fenolik 500

Design points above predicted value
Design points below predicted value 480
374,123 482,677 460
440
total fenolik

X1 = A: teh putih 420

X2 = B: teh hijau 400
X3 = C: teh hitam
380
360

A (1)
B (0)
C (1)

C (0)
A (0)

B (1)