全部版块 我的主页
论坛 数据科学与人工智能 数据分析与数据科学 SAS专版
2050 2
2010-01-08
I'm analyzing a data set (given below) using the SAS codes:
proc glm data=par3;
class S P;
model y = V*S V*P / solution;
estimate 'C - L at V=0.05 & P=large' V*S 0.05 -0.05 V*P 0.05;
estimate 'C - L at V=0.05 & P=small' V*S 0.05 -0.05;
run;

and would like to compare mean Y at each of four settings of S*P. Class variable S takes two levels: C vs L, class variable P takes two levels: large and small. But SAS gave a error message:
C - L at V=0.05 & P=large is not estimable.

Experts, any useful advice would be highly appreciated. Thanks in advance!!!

Data:

y S P V
29.18 C large 0.042954
29.94 C large 0.042954
32.75 C large 0.042954
22.38 C large 0.026264
20.56 C large 0.026264
11.39 C large 0.008532
8.89 C large 0.008532
9.58 C large 0.008532
25.64 C large 0.039110
25.90 C large 0.039840
25.59 C large 0.039840
28.25 C large 0.039840
66.95 C large 0.084107
64.24 C large 0.084107
62.76 C large 0.084107
39.20 C small 0.044582
35.61 C small 0.044582
31.22 C small 0.044582
21.27 C small 0.027473
23.62 C small 0.027473
22.44 C small 0.027473
12.29 C small 0.009818
14.00 C small 0.009818
9.90 C small 0.009818
28.44 C small 0.039110
29.61 C small 0.039110
59.56 C small 0.081520
67.51 C small 0.081520
58.74 C small 0.081520
30.49 L large 0.042954
30.09 L large 0.042954
29.04 L large 0.042954
19.43 L large 0.026264
20.11 L large 0.026264
21.65 L large 0.026264
10.41 L large 0.008532
8.93 L large 0.008532
10.28 L large 0.008532
9.41 L large 0.009818
58.46 L large 0.084107
54.46 L large 0.084107
51.88 L large 0.084107
34.08 L small 0.044582
32.47 L small 0.044582
29.18 L small 0.044582
21.67 L small 0.027473
19.55 L small 0.027473
18.32 L small 0.027473
21.61 L small 0.027473
20.55 L small 0.027473
20.44 L small 0.027473
10.15 L small 0.009818
9.64 L small 0.009818
60.58 L small 0.081520
63.96 L small 0.081520
56.77 L small 0.081520


Partial results:

The GLM Procedure
Dependent Variable: y
Sum of
Source DF Squares Mean Square F Value Pr > F
Model 3 17040.43849 5680.14616 783.96 <.0001
Error 52 376.76153 7.24541
Corrected Total 55 17417.20002

R-Square Coeff Var Root MSE y Mean
0.978368 8.913964 2.691731 30.19679

Source DF Type I SS Mean Square F Value Pr > F
V*S 2 16983.83189 8491.91595 1172.04 <.0001
V*P 1 56.60660 56.60660 7.81 0.0073

Source DF Type III SS Mean Square F Value Pr > F
V*S 1 85.20517671 85.20517671 11.76 0.0012
V*P 1 56.60659538 56.60659538 7.81 0.0073

Standard
Parameter Estimate Error t Value Pr > |t|
Intercept 2.6310648 0.67351512 3.91 0.0003
V*S C 733.3991228 B 18.03198481 40.67 <.0001
V*S L 681.2693162 B 18.00808459 37.83 <.0001
V*P large -42.4603440 B 15.19083328 -2.80 0.0073
V*P small 0.0000000 B . . .


二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

全部回复
2010-1-8 07:09:45
1# jennyli1346

复制代码
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

2010-1-8 08:03:58
jingju11,
Thank you very much! you're a SUPER genius! I guess that your codes worked as the coefficient corresponding to P=small is zero.

Jenny
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

相关推荐
栏目导航
热门文章
推荐文章

说点什么

分享

扫码加好友,拉您进群
各岗位、行业、专业交流群