build bridge to blavaan

[jpritikin]

Blavaan would be a good way to start exploring Bayesian MCMC approaches.

[tbates]

Sounds like this uses lavaan syntax to spec Bayes models + can edit JAGS code.

We might just modify RAM to write JAGS code?

Merkle, Rosseel (2018). blavaan: Bayesian Structural Equation Models via Parameter Expansion https://www.jstatsoft.org/article/view/v085i04

This article describes blavaan, an R package for estimating Bayesian structural equation models (SEMs) via JAGS and for summarizing the results. It also describes a novel parameter expansion approach for estimating specific types of models with residual covariances, which facilitates estimation of these models in JAGS. The methodology and software are intended to provide users with a general means of estimating Bayesian SEMs, both classical and novel, in a straightforward fashion. Users can estimate Bayesian versions of classical SEMs with lavaan syntax, they can obtain state-of-the-art Bayesian fit measures associated with the models, and they can export JAGS code to modify the SEMs as desired. These features and more are illustrated by example, and the parameter expansion approach is explained in detail.

[jpritikin]

Sounds like this uses lavaan syntax to spec Bayes models

Yes

We might just modify RAM to write JAGS code?

Sure, but I think it would be easier to leverage Ed Merkle's work by using lavaan as an intermediate representation.

[mcneale]

Are there limits to the set of models that Lavaan can fit, compared to OpenMx? Does it do mixture distributions, for example? From what I see online, it's a planned future feature but not currently implemented http://lavaan.ugent.be/development.html So it seems to me that going via Lavaan could lose significant functionality (and put OpenMx at their mercy re: code changes that break what we do).