Dear experts,
I’m running a DCM with 4 nodes and a rather large model space and doing family selection/averaging. I’ve got 12 families in total, but they vary in size, with most having between 2 and 8 members, but one having a whopping 32, as it is the most complex.
Is this a problem? Am I biasing things towards the larger family? I’m asking because, it wins the BMS clearly, but compared to other families it has few good fitting models, just a bunch of low-average models. A couple of other families tend to have only one good fitting model, which I would imagine drags the average down.
Any help much appreciated,
Haakon
To unsubscribe from the SPM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=SPM&A=1
|