24小时热门版块排行榜

返回列表

wangyx0411702

铁杆木虫 (正式写手)

应助: 3 (幼儿园)
金币: 5243.1
散金: 50
红花: 2
帖子: 565
在线: 223.4小时
虫号: 480112
注册: 2007-12-15
专业: 食品科学

[交流] design expert软件使用求助已有1人参与

师兄师姐们，请教一个优化类问题。我在文献中看到用design expert软件进行Plackett–Burman结果分析时，有对各实验因素进行Standardized effects, F-value, p-value, t-value, Sum of squares, Mean square等分析的数据表，但是我用软件分析后却得不全。只有对响应值有显著影响的因素有这些数据，其它没有显著影响且被软件自动去除的因素，软件却没有给出。我的操作是这样的：
1、我的P-B实验数据输入后，在Analysis板块下，点击Effects后，再点击左侧Effects Tool小框中的Effects list，随后在右边对话框中Selection下选择Backgward, 其后的Alpha Out 项自动给出为0.1000, 最后的Order下选择Main effect。
2、然后，再点击ANOVA，出现的结果框中显示出：
         Coefficient t for H0
Removed          Estimate Coeff=0    Prob > |t| R-Squared MSE
  F-Urea                -0.00169    0.14687 0.1000 1.000
  A-Glucose -0.00194    -1.15252    0.455 0.997435       3.41E-05
  L-FeSO4                   -0.00518    -2.85008    0.1042 0.994029       3.97E-05
  C-Sucrose -0.00544    -1.62793    0.202 0.969776       0.000134
  H-NH4Cl                   -0.00561    -1.41422    0.2302 0.943077       0.000189
  D-Peptone 0.006733    1.54841    0.1822 0.914615       0.000227
  E-Ammonium sulfate0.00838    1.735416  0.1334 0.873671       0.00028

ANOVA for selected factorial model
Analysis of variance table [Partial sum of squares - Type III]
Source          Sum of                Mean                         p-value
                     Squares       df       Square F Value    Prob > F
Model             0.011 4 2.69E-03 7.47 0.0114 significant
B-Dextrin             2.43E-03 1 2.43E-03 6.74 0.0356
G-Yeast extract  2.94E-03 1 2.94E-03 8.17 0.0244
J- K2HPO4       2.19E-03 1 2.19E-03 6.09 0.043
K- MgSO4          3.20E-03 1 3.20E-03 8.89 0.0205
Residual          2.52E-03 7 3.60E-04
Cor Total          0.013 11

The Model F-value of 7.47 implies the model is significant.  There is only a 1.14% 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 B, G, J, K 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.
      Coefficient       Standard 95% CI 95% CI
Factor          Estimate df Error Low High       VIF
Intercept          0.039747  1 0.005479 0.026792 0.052702
B-Dextrin          -0.01423 1 0.005479 -0.02718 -0.00127       1
G-Yeast extract 0.01566 1 0.005479 0.002705 0.028615       1
J- K2HPO4       0.01352 1 0.005479 0.000566 0.026476       1
K- MgSO4             0.01633 1 0.005479 0.00338 0.029289       1
随后就是对响应给出的方程了。这个软件从一无所知到现在知之不多，已经折腾我两个月了，各位专家级师兄师姐，请不吝帮我分析一下好吗？再次成分感激，谢谢您！！！

回复此楼