# Determine intrinsic aliasing in linear mixed model

#### downthesun01

##### New Member
A bunch of us are studying for an upcoming actuarial exam and none of us understands item II. doesn't result in intrinsic aliasing.

We've calculated the V = ZDZ' +R_1 matrix and as far as we can tell, there's aliasing. But, maybe there's something about this all that we're not understanding.
Linear Mixed Models: A Practical Guide Using Statistical Software (Second Edition) by West et al is the source for this section of the exam, but the question the ASM study manual.
Any help with this would be appreciated. Thanks.

Images of the problem, solution and a couple pages of the case study are included below, but for those of you having trouble with the images I've typed out some of that:

This is what statement II says. The given solution says it's false because the model has two random effects.
II. states that using a compound symmetry structure for the residual in Model 1 would result in intrinsic aliasing.

Model 1's specifications are: Fixed effects: GRADE, SEX Random effects: INTERCEPT, SEX for CLASS

In this case study, test scores of students are observed over 4 grades, numbered 0, 1, 2, and 3. There are 70 students in 10 classes. Each student stays in the same class for all 4 grades.
Data Fields: SCORE - score on the exam (response)
STUDENT - identifier of student
GRADE - grade of the student (0, 1, 2, 3) SEX - sex of the student (0, 1) CLASS - Identifier of student's class   