StructuralEquationModels.jl

StructuralEquationModels.jl copied to clipboard

gather collection of complicated SEM models to guide users in their specification

2

# LGCM ```julia using StenoGraphs, StructuralEquationModels ts = 1:100 obs = Symbol.("vf_t", lpad.(ts, 3, "0")) @declare_nodes i s # same as i, s = Node.((:i, :s)) m = Node(Symbol("1")) graph...

aaronpeikert

Is there a way to specify a model without data? For example, if I specify a population model in a simulation, I never want to fit this but directly simulate...

brandmaier

Numerical precision of model-implied covariance matrix

3

I encountered a problem with the model-implied covariance matrix of a SEM in SEM.jl. The model is a standard linear latent growth curve model fitted on data perfectly simulated from...

brandmaier

We should have a way to display standardized estimates (based on a parameter table?)

brandmaier

Alexey is not listed as co-author in the metadata of the preprint on Psyarxiv.

brandmaier

CFI

1

We need CFI as a standard fit measure because RMSEA and CFI together are the de facto reporting standard. However, it needs the fit of a null model, which usually...

brandmaier

Does the package support definition variables as in OpenMx? If not, I strongly vote for including this in both the package and Stenograph. If this is not yet, implemented, it...

brandmaier

Maximilian-Stefan-Ernst

(in)equality constraints via StenoGraphs specification

3

Currently, (in)equality constraints require some manual labor and are relatively error-prone. This is from the documentation: ```{julia} parind[:y3y7] # 29 parind[:y8y4] # 30 # θ[29] + θ[30] - 1 =...

brandmaier

Add a warning when meanstructure=false and means are specified?

1

Currently, if I do ``` model = Sem( specification = partable, data = data, meanstructure = false ) ``` and the partable has a mean structure specified, this is just...

brandmaier