q**t
发帖数: 36 | 1 I met a problem to test a nested design with 4 levels. There are one fixed
effect x1.
random effects are nested A, B(A), C( B*A)
For each subgroup C, there are D repeated measurements.
I need to estimated the variance within each subgroup C. Also
I want to test for each subgroup B, the variance within subgroup C
heterogeneous or not?
To estimate the variance within C, here is my code:
*** I try to test the heterogeneity using covtest;
proc mixed data= data4analysis cl covtest ;
class A B C D;
model Y = X1/ss3 solution ddfm =bw;
random intercept/ subject =A;
random intercept/ subject =B(A);
random intercept/ subject =C(B*A);
repeated D / subject =C(B*A) group =B;
run;
Is the estimation correct? Why can't get any variance likely ratio test as
when use repeated statement only and delete all the random effects? As
following:
proc mixed data= data4analysis cl covtest ;
class A B C D;
model Y = X1/ss3 solution ddfm =bw;
repeated D / subject =C(B*A) group =B;
run;
The estimations are different in this way.
Can anyone help me on this? Thanks a lot! | s******0
发帖数: 1269 | 2 第一个,没明白你的design structure.你想说的是有random*random interaction?
第二个,你觉得这2段code想要做的东西一样吗?不一样的话结果为什么要一样呢? | q**t
发帖数: 36 | 3 Design structure is like this:
level A : 1 2 .... 5
level B : a b c d l m
level C: x y z
level D repeat: 1 1 1
2 2 2
3 3 3
So all levels A, B, C are random and D is repeat
I thought the first code is more suitable for estimation of the variance
within each subgroup C ,for each group B. I might messed up here.
2nd one is not assume that A, B, C are random, but it can give me a covtest
result of the variance homogeneity across the groups B. |
|
