问题：在做联合分析法用SAS设计问卷时，对2个2水平（x1, x3），2个3水平(x2, x4)，做所有的main effect 和 two-way interaction。
SAS程序（参考资料：Multinomial Logit,Discrete Choice Modeling An Introduction to Designing Choice Experiments,and Collecting, Processing, and Analyzing Choice Data with the SAS System，见附件）

%mktdes(factors=x1 x3=3 x2 x4=2,interact=x1*x2 x1*x3 x1*x4 x2*x3 x2*x4 x3*x4,procopts=seed=7654321);
%choiceff(data=final, model=class(x1-x4 x1*x2 x1*x3 x1*x4 x2*x3 x2*x4 x3*x4), nsets=12,seed=9999, flags=f1-f3, beta=zero);
proc print;
by set;
id set;run;

结果：在最后面（结果删略了一部分，字数太多了，不能发帖）

问题（结果中的红色部分）：
1、请问程序有没有问题呢？interaction是这样估算的吗？模型中的变量是这样设置的吗？（model=class(x1-x4 x1*x2 x1*x3 x1*x4 x2*x3 x2*x4 x3*x4）
2、结果中有两个D-efficiency（93.8478，0.6069），请问分别是什么意思呢？
3、Parameters 19，Maximum Parameters 24。因为24>19+1，所以可以估算出结果吗？
4、很抱歉@了坛友，刚刚接触SAS和conjoint analysis，所以很多问题都不懂，谢谢帮忙解答哦。


The OPTEX Procedure

Class Level Information

ClassLevels -Values-

x1 3 1 2 3
x3 3 1 2 3
x2 2 1 2
x4 2 1 2

The OPTEX Procedure

Average
Prediction
Design Standard
Number D-Efficiency A-Efficiency G-Efficiency Error
------------------------------------------------------------------------
1 93.8478 86.5979 72.0082 0.8774
2 93.8478 86.5979 72.0082 0.8774
3 93.8478 86.5979 72.0082 0.8774
4 93.8478 86.5979 72.0082 0.8774
5 93.8478 86.5979 72.0082 0.8774

n Name Beta Label

1 x11 0 x1 1
2 x12 0 x1 2
3 x21 0 x2 1
4 x31 0 x3 1
5 x32 0 x3 2
6 x41 0 x4 1
7 x11x21 0 x1 1 * x2 1
8 x12x21 0 x1 2 * x2 1
9 x11x31 0 x1 1 * x3 1
10 x11x32 0 x1 1 * x3 2
11 x12x31 0 x1 2 * x3 1
12 x12x32 0 x1 2 * x3 2
13 x11x41 0 x1 1 * x4 1
14 x12x41 0 x1 2 * x4 1
15 x21x31 0 x2 1 * x3 1
16 x21x32 0 x2 1 * x3 2
17 x21x41 0 x2 1 * x4 1
18 x31x41 0 x3 1 * x4 1
19 x32x41 0 x3 2 * x4 1


Design IterationD-Efficiency D-Error
----------------------------------------------
1 0 0 .
1 0.566252 1.765997
2 0.598419 1.671069
3 0.605172 1.652423
4 0.606231 1.649536
Design IterationD-Efficiency D-Error
----------------------------------------------
2 0 0.329156 3.038072
1 0.596069 1.677658
2 0.605168 1.652433
3 0.606911 1.647687
Final Results

Design 2
Choice Sets 12
Alternatives 3
Parameters 19
Maximum Parameters 24
D-Efficiency 0.6069
D-Error 1.6477



Variable Standard
n Name Label Variance DF Error
1 x11 x1 1 3.83632 1 1.95865
2 x12 x1 2 3.63944 1 1.90773
3 x21 x2 1 2.60984 1 1.61550
4 x31 x3 1 4.12999 1 2.03224
5 x32 x3 2 4.05790 1 2.01442
6 x41 x4 1 2.49144 1 1.57843
7 x11x21 x1 1 * x2 1 3.08844 1 1.75740
8 x12x21 x1 2 * x2 1 3.78379 1 1.94520
9 x11x31 x1 1 * x3 1 4.20355 1 2.05026
10 x11x32 x1 1 * x3 2 4.34657 1 2.08484
11 x12x31 x1 2 * x3 1 4.04728 1 2.01179
12 x12x32 x1 2 * x3 2 3.73853 1 1.93353
13 x11x41 x1 1 * x4 1 2.71513 1 1.64776
14 x12x41 x1 2 * x4 1 2.92248 1 1.70953
15 x21x31 x2 1 * x3 1 3.34435 1 1.82876
16 x21x32 x2 1 * x3 2 3.02975 1 1.74062
17 x21x41 x2 1 * x4 1 2.56929 1 1.60290
18 x31x41 x3 1 * x4 1 3.70047 1 1.92366
19 x32x41 x3 2 * x4 1 3.68237 1 1.91895
==
19
Set Design Efficiency Index Prob n f1 f2 f3 x1 x2 x3 x4
1 2 0.60691 2 0.33333 37 1 1 1 1 1 1 2
2 0.60691 13 0.33333 38 1 1 1 2 1 1 1
2 0.60691 3 0.33333 39 1 1 1 1 1 2 1
2 2 0.60691 5 0.33333 40 1 1 1 1 1 3 1
2 0.60691 28 0.33333 41 1 1 1 3 1 2 2
2 0.60691 24 0.33333 42 1 1 1 2 2 3 2
3 2 0.60691 24 0.33333 43 1 1 1 2 2 3 2
2 0.60691 9 0.33333 44 1 1 1 1 2 2 1
2 0.60691 21 0.33333 45 1 1 1 2 2 2 1