Source DF Sum of Squares Mean Square F Value Pr > F
Model 8 5.79920000 0.72490000 . .
Error 0 . .
Corrected Total 8 5.79920000
R-Square C.V. Root MSE Y Mean
1.000000 0 5.83666667
Source DF Anova SS Mean Square F Value Pr > F
X1 2 0.41086667 0.20543333 . .
X2 2 0.10806667 0.05403333 . .
X3 2 4.44740000 2.22370000 . .
X4 2 0.83286667 0.41643333 . .作者: shiyiming 时间: 2008-11-3 06:10 标题: Re: 求助！四因素三水平的正交实验编程 no enough observations maybe.作者: shiyiming 时间: 2008-11-3 09:04 标题: Re: 求助！四因素三水平的正交实验编程 这样说，四因素三水平的试验用l9（34）的表不行咯作者: shiyiming 时间: 2008-11-3 14:14 标题: Re: 求助！四因素三水平的正交实验编程 Your design table is OK (9 is the number for 3^4), but you don't have adequate # of measurements for the proposed model.
Anyway, don't trust me, I'm still dangling at the bottom of the food chain!作者: shiyiming 时间: 2008-11-3 20:34 标题: Re: 求助！四因素三水平的正交实验编程 So could you tell me how to Analysis it?
and what to write作者: shiyiming 时间: 2008-11-4 01:59 标题: Re: 求助！四因素三水平的正交实验编程 it's beyond my capacity. i don't know how to fit the model either.
my hunch is that you don't have enough data for the proposed analysis. the orthogonal design is a magic to reduce the # of runs, but several repetitions per run may still be desired to fit the model.作者: shiyiming 时间: 2008-11-4 08:19 标题: Re: 求助！四因素三水平的正交实验编程 oh,Isee.You are wonderful! <!-- s:D --><img src="{SMILIES_PATH}/icon_biggrin.gif" alt=":D" title="Very Happy" /><!-- s:D -->
thank you very much!作者: shiyiming 时间: 2008-11-9 14:04 标题: Re: 求助！四因素三水平的正交实验编程 四个因子三个水平，每个因子要有2个自由度，所以因子总共就要有8个自由度，而你总共只有9个观测值，即只有8个自由度可以分配，所以8-8＝0就没有自由度给 残差（Error）了。