1. Run "IntroSEM-BivariateRaw.R" in trunk.models/passing
2. Send the output to omxGraphviz.
Despite this being a RAM model its type is not "MxRAMModel"
# (which makes omxGraphviz() terminate with error )
RAM + raw data is converted into a FIML model. If we want to change the specification to return a RAM model as output instead of a FIML model, we have to decide where to put the new 'covariance' and 'means' algebras that are generated by the FIML model, and are explicitly used in several of our test cases.
I guess this is a RAMFIML model then - it still meets all the crtieria of a RAM model in terms of what the matrices do?
Perhaps RAMFIML as a type, or else alter graphviz to detect that this model is OK for its purposes?
The RAM + raw data transformation has been fixed. See http://openmx.psyc.virginia.edu/2011/05/phasing-out-ram-fiml-transformation