Try "Causality: Models, reasoning, and inference" by Judea Pearl. As I
understand it, pearl's approach combines the structural equation modeling
and Bayesian Inference network. And Pearl favors over-identified models.
When there are many ways to fit the data and the model but you found a
unique solution, a causal inference can be made.
The building block is the latent factor model--the causal relationship
between the theoretical construct and the observed items. According to
Borsboom, D., Mellenbergh, G. J., & van Heerden, the logic is like this:
operationalism and the latent construct theory are fundamentally
incompatible. If a latent construct is just for operational convenience,
then there should be a distinct latent factor for every single test
researchers construct. However, since it is assumed that observed items
that are loaded into a factor constitute a single dimension, theoretical
construct are causally responsible for observed phenomena.
Chong Ho (Alex) Yu, Ph.D.
On Wed, 21 Feb 2001, G. Anthony Reina wrote:
> Is there a test that variable X causes variable Y?
>
> I was under the impression that the best statistics could do was
> correlation not causation. In order to prove causation, one would have
> to know the specific mechanism whereby X could cause Y and possibly vary
> the input X to see if Y changed accordingly.
>
> However, I've seen some papers on a method called 'directed coherence'
> which uses something called Granger causality. I think the basic gist is
> that the 'directed coherence' is the probability of predicting something
> about Y given you know something about X.
>
> Has anyone run across this?
>
> Thanks.
> -Tony Reina
>
>
>
>
> =================================================================
> Instructions for joining and leaving this list and remarks about
> the problem of INAPPROPRIATE MESSAGES are available at
> http://jse.stat.ncsu.edu/
> =================================================================
>
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
