14.03.2014 Views

Modeling and Multivariate Methods - SAS

Modeling and Multivariate Methods - SAS

Modeling and Multivariate Methods - SAS

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Appendix A Statistical Details 665<br />

The Factor Models<br />

JMP <strong>and</strong> GLM Hypotheses<br />

GLM works differently than JMP <strong>and</strong> produces different hypothesis tests in situations where there are<br />

missing cells. In particular, GLM does not recognize any difference between a nesting <strong>and</strong> a crossing in an<br />

effect, but JMP does. Suppose that you have a three-layer nesting of A, B(A), <strong>and</strong> C(A B) with different<br />

numbers of levels as you go down the nested design.<br />

Table A.10 on page 665, shows the test of the main effect A in terms of the GLM parameters. The first set<br />

of columns is the test done by JMP. The second set of columns is the test done by GLM Type IV. The third<br />

set of columns is the test equivalent to that by JMP; it is the first two columns that have been multiplied by<br />

a matrix<br />

21<br />

12<br />

to be comparable to the GLM test. The last set of columns is the GLM Type III test. The difference is in<br />

how the test distributes across the containing effects. In JMP, it seems more top-down hierarchical. In GLM<br />

Type IV, the test seems more bottom-up. In practice, the test statistics are often similar.<br />

Table A.10 Comparison of GLM <strong>and</strong> JMP Hypotheses<br />

Parameter JMP Test for A GLM-IV Test for A JMP Rotated Test GLM-III Test for A<br />

u 0 0 0 0 0 0 0 0<br />

a1 0.6667 -0.3333 1 0 1 0 1 0<br />

a2 –0.3333 0.6667 0 1 0 1 0 1<br />

a3 –0.3333 -0.3333 -1 -1 -1 -1 -1 -1<br />

a1b1 0.1667 -0.0833 0.2222 0 0.25 0 0.2424 0<br />

a1b2 0.1667 -0.0833 0.3333 0 0.25 0 0.2727 0<br />

a1b3 0.1667 -0.0833 0.2222 0 0.25 0 0.2424 0<br />

a1b4 0.1667 -0.0833 0.2222 0 0.25 0 0.2424 0<br />

a2b1 -0.1667 0.3333 0 0.5 0 0.5 0 .5<br />

a2b2 -0.1667 0.3333 0 0.5 0 0.5 0 .5<br />

a3b1 -0.1111 -0.1111 -0.3333 -0.3333 -0.3333 -0.3333 -0.3333 -0.3333<br />

a3b2 -0.1111 -0.1111 -0.3333 -0.3333 -0.3333 -0.3333 -0.3333 -0.3333

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!