Hello, a customer discovered a new problem. When he models the same model in both JMP and SAS it does not give the same estimates. JMP SAS Code: proc mixed data = testData; class F1 F2 F3 ; model V1 = F2(F1) / DDFM=KENWARDROGER cl; random F3; lsmeans F2(F1) / cl; run; Do you know why this happens? Thanks in advance, Valerie

nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Follow up question: Linear mixed models JMP 10 versus SAS 9.2

Valerie_Nedbal

Hello,

a customer discovered a new problem. When he models the same model in both JMP and SAS it does not give the same estimates.

JMP

SAS

Code:

proc mixed data = testData;

class

;

model V1 = F2(F1) / DDFM=KENWARDROGER cl;

random F3;

lsmeans F2(F1) / cl;

run;

Do you know why this happens?

Thanks in advance,

Valerie

Accepted answers

jiancao

It seems the way JMP calculates the LS Means for the nested effect F2[F1] is different from SAS. I've also noticed that out of six levels of F2[F1], JMP produces three estimates (NParm=3 and DF=3), while you will get five estimates if you run the SAS code. As Jeff suggests, Tech Support should be able to shed light why the difference.

All comments

Jeff_Perkinson

It might be most expeditious to have the user contact tech support directly (support@jmp.com).

They are best positioned to investigate and explain the difference here.

-Jeff

jiancao

Valerie_Nedbal

OK will do, I will let you know the outcome.

Best,

Valerie