ElMaestro ★★★ Denmark, 2020-08-14 22:54 (1545 d 03:17 ago) Posting: # 21882 Views: 4,473 |
|
Hi all, can someone post REML results for dataset I of EMA? For example a WNL output and/or SAS output of variance components and fixed effects. I will be using it to verfiy my own scripts in R, or at least do some direct comparison. Thanks in advance if possible. — Pass or fail! ElMaestro |
PharmCat ★ Russia, 2020-08-15 00:09 (1545 d 02:02 ago) @ ElMaestro Posting: # 21883 Views: 3,632 |
|
❝ Hi all, Hi! SPSS Code:
MIXED var1 BY sequence period formulation Information Criteria
-2 Restricted Log Likelihood 530.145 Estimates of Covariance Parametersa
Random Effect Covariance Structure (G) Pairwise Comparisons Julia code:
using ReplicateBE |
ElMaestro ★★★ Denmark, 2020-08-15 00:34 (1545 d 01:37 ago) @ PharmCat Posting: # 21884 Views: 3,621 |
|
Thanks PharmCat - again, your help here is very valuable to me. — Pass or fail! ElMaestro |
Helmut ★★★ Vienna, Austria, 2020-08-15 17:00 (1544 d 09:11 ago) @ ElMaestro Posting: # 21885 Views: 3,609 |
|
Hi ElMaestro, second opinion: Banded No-Diagonal factor Analytic (f=2)
Banded No-Diagonal factor Analytic (f=1)
Heterogenous Compound Symmetry
A correlation >1 is funny. Compound Symmetry
Unstructured
EMA Method A
EMA Method B (residual df like SAS DDFM=CONTAIN)
EMA Method B (Satterthwaite’s df)
— Dif-tor heh smusma 🖖🏼 Довге життя Україна! Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
PharmCat ★ Russia, 2020-08-16 02:57 (1543 d 23:14 ago) @ Helmut Posting: # 21886 Views: 3,591 |
|
Hi Helmut! ❝ -2REML LL 504.28342 I suppose Phoenix REML includes/exludes some constant. ❝ A correlation >1 is funny. I think CSH is most fair structure. But if otimize rho without limitaion or link function result cab be very srange |
ElMaestro ★★★ Denmark, 2020-08-20 00:50 (1540 d 01:21 ago) @ Helmut Posting: # 21891 Views: 3,428 |
|
Hi Hötzi, thanks for WNL outputs with various covariance structures. ❝ A correlation >1 is funny. In a way you can compare it with the CoA of a reference standard which is labeled with purity = 101.2 %. It does not contain 101.2% - I guarantee it - but that may stil be the most viable estimate. It is all about how we estimate / measure it. We are not guaranteed to get the same model maximum likelihood if you keep the rho controlled between (exactly) -1 to (exactly) 1 with any iterative method regardless of your tolerance setting on the x- or y-scale. — Pass or fail! ElMaestro |
Helmut ★★★ Vienna, Austria, 2020-08-20 03:05 (1539 d 23:05 ago) @ ElMaestro Posting: # 21892 Views: 3,459 |
|
Hi ElMaestro, ❝ ❝ A correlation >1 is funny. ❝ ❝ In a way you can compare it with the CoA of a reference standard which is labeled with purity = 101.2 %. It does not contain 101.2% - I guarantee it - but that may stil be the most viable estimate. That’s another cup of tea. I guarantee that it can contain exactly 101.2% indeed – regardless what the CoA says. ❝ It is all about how we estimate / measure it. We are not guaranteed to get the same model maximum likelihood if you keep the rho controlled between (exactly) -1 to (exactly) 1 with any iterative method regardless of your tolerance setting on the x- or y-scale. Here you are right despite some weirdos laments.* The largest correlation in my 500 simulated data sets was for #168 with 7.285 (‼)…
— Dif-tor heh smusma 🖖🏼 Довге життя Україна! Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |