一个模型拟合的结果，请帮忙分析，谢谢！
用的Design expert做的2个因素的mixture设计，结果发现方程显著，但失拟不显著，但后面软件又给出了This model can be used to navigate the design space.不明白什么意思，请高手指点下，下面是输出的结果，多谢！
Response 1 R1
ANOVA for Mixture Quadratic Model
Mixture Component Coding is L_Pseudo.
Analysis of variance table
Sum of Mean F p-value
Source Squares df Square Value Prob > F
Model 0.022 2 0.011 70.37 < 0.0001 significant
Linear Mixture 0.012 1 0.012 74.29 < 0.0001
AB 0.010 1 0.010 66.46 < 0.0001
Residual 1.552E-003 10 1.552E-004
Lack of Fit 1.346E-003 2 6.730E-004 26.09 0.0003 significant
Pure Error 2.064E-004 8 2.580E-005
Cor Total 0.023 12
The Model F-value of 70.37 implies the model is significant. There is only
a 0.01% chance that a "Model F-Value" this large could occur due to noise.
Values of "Prob > F" less than 0.0500 indicate model terms are significant.
In this case Linear Mixture Components, AB 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.
The "Lack of Fit F-value" of 26.09 implies the Lack of Fit is significant. There is only a
0.03% chance that a "Lack of Fit F-value" this large could occur due to noise.
Significant lack of fit is bad -- we want the model to fit.
Std. Dev. 0.012 R-Squared 0.9337
Mean 0.080 Adj R-Squared 0.9204
C.V. % 15.66 Pred R-Squared 0.8970
PRESS 2.411E-003 Adeq Precision 17.174
The "Pred R-Squared" of 0.8970 is in reasonable agreement with the "Adj R-Squared" of 0.9204.
"Adeq Precision" measures the signal to noise ratio. A ratio greater than 4 is desirable. Your
ratio of 17.174 indicates an adequate signal. This model can be used to navigate the design space.
Response 1 R1
ANOVA for Mixture Quadratic Model
Mixture Component Coding is L_Pseudo.
Analysis of variance table
Sum of Mean F p-value
Source Squares df Square Value Prob > F
Model 0.022 2 0.011 70.37 < 0.0001 significant
Linear Mixture 0.012 1 0.012 74.29 < 0.0001
AB 0.010 1 0.010 66.46 < 0.0001
Residual 1.552E-003 10 1.552E-004
Lack of Fit 1.346E-003 2 6.730E-004 26.09 0.0003 significant
Pure Error 2.064E-004 8 2.580E-005
Cor Total 0.023 12
The Model F-value of 70.37 implies the model is significant. There is only
a 0.01% chance that a "Model F-Value" this large could occur due to noise.
Values of "Prob > F" less than 0.0500 indicate model terms are significant.
In this case Linear Mixture Components, AB 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.
The "Lack of Fit F-value" of 26.09 implies the Lack of Fit is significant. There is only a
0.03% chance that a "Lack of Fit F-value" this large could occur due to noise.
Significant lack of fit is bad -- we want the model to fit.
Std. Dev. 0.012 R-Squared 0.9337
Mean 0.080 Adj R-Squared 0.9204
C.V. % 15.66 Pred R-Squared 0.8970
PRESS 2.411E-003 Adeq Precision 17.174
The "Pred R-Squared" of 0.8970 is in reasonable agreement with the "Adj R-Squared" of 0.9204.
"Adeq Precision" measures the signal to noise ratio. A ratio greater than 4 is desirable. Your
ratio of 17.174 indicates an adequate signal. This model can be used to navigate the design space.
没有找到相关结果
已邀请:
RSS Feed
1 个回复
wang168 (威望:20) (台湾 台湾) 咨询业 咨询顾问
赞同来自:
This model can be used to navigate the design space.
有点好奇问楼主，做 2 个因素 Mixture 设计，一般 Mixture 设计是 3 个因素以上，LZ 如何办到 ?