SEM and Dichotomous Variables
David Liu writes:
In Statistics and Causal Inference: A Review (Pearl 2003), it was said 'the bulk of SEM methodology was developed for linear analysis, and until recently, no comparable methodology has been devised to extend its capabilities to models involving dichotomous variables or nonlinear dependencies.' Is it true by now?
To the best of my knowledge, it is still true. The non-parametric SEM techniques expounded in my book do capture, of course, dichotomous variables andother nonlinear dependencies. However, I am not aware of methodologies that exploit to advantage the unique features of dichotomous variable to facilitate causal analysis.
Comment by judea — August 6, 2007 @ 5:10 pm
Quality articles or reviews is the important to interest the viewers
to visit the website, that’s what this site is providing.
Comment by Terese — August 22, 2014 @ 4:25 am