Dear OpenMx experts,

I'm now working on a twin dataset with zygosity diagnosis is not exact, i.e. I only have one variable indicating the probability of one twin pair being MZ or DZ. In Mx we can "specify" this variable and "weight" it when estimating the covariance matrix. Dr. Neale has discussed this issue and provided the Mx script file here:

ftp://ftp.vcu.edu/pub/mx/examples/zygosity

The key commands are below:

G2: All twin pairs

Data NInput_vars=3 Nmodel=2

Labels t1 t2 pmz

Rectangular File=sim1.mz

Select t1 t2 pmz ;

Definition pmz ;

Matrices= Group 1

P Full 1 1

End Matrices;

Specify P pmz

Means N;

Covariances A+C+D+E | A+C+D _

A+C+D | A+C+D+E _

A+C+D+E | H@A+C+Q@D _

H@A+C+Q@D | A+C+D+E ;

Weights P_I-P;

My question is how we can do the same analysis in OpenMx? Any help will be greatly appreciated.

Best,

Shaoyong

This is part of the test suite, whose scripts are here:

http://openmx.psyc.virginia.edu/svn/trunk/models/passing/

specifically, take a look at:

http://openmx.psyc.virginia.edu/svn/trunk/models/passing/Acemix.R

and

http://openmx.psyc.virginia.edu/svn/trunk/models/passing/Acemix2.R

Thanks so much. This is realy helpful!