Dear Marco,
this may be a bit late, but regarding your question on second-level model assessment (from November 2017, see below):
I have added functionality for this to the MACS toolbox last week [1] and also submitted a validation analysis as an OHBM abstract today [2,3].
If you have any questions, just let me know.
[1] https://github.com/JoramSoch/MACS/commit/c64029cc9775b6577d98312123a0c6b8bbb4f5ed
[2] https://ww5.aievolution.com/hbm1901/index.cfm?do=abs.viewAbs&subView=1&abs=4707
[3] https://www.dropbox.com/s/k0w9pd1p1awzsnf/Abstract_2018_12_19.pdf?dl=0
Cheers
Joram
----- Original Message -----
From: "Marco Marco Phil Caviezel" <[log in to unmask]>
To: [log in to unmask]
Sent: Thursday, November 16, 2017 5:50:23 PM
Subject: [SPM] model-fit of 2nd level GLM
Dear SPMers,
I’m interested how many regressors of no-interest (e.g. age, gender etc.) I should put into my 2nd level GLM as covariates and how they affect the GLM.
Do you know a method to get the model-fit (e.g. AIC, BIC, etc.) of a 2nd level GLM, so I can compare two or more different models?
Using the MACS Toolbox ("MACS – a new SPM toolbox for model assessment, comparison and selection", 2017 by Joram Soch, Carsten Allefeld) I found a way to compare GLMs on 1st level, but to my knowledge, this doesn’t work on 2nd level.
I appreciate any help and ideas!
Best,
Marco
Marco Caviezel
PhD Student
University of Basel
|