v0.5.0 #293
Copy link
Copy link
Open
Description
-
StenoGraphshas to be loaded manually #289 - MethodError: no method matching SemObservedData #283
-
sem_fit()was deprecated forfit()? #267 - CFI #272
- Explicitely import types that are used by package extensions #292
- Add update_partable! method to allow to pass a vector of pairs instead of dict #227
- Misc. doc fixes #251
- se for WLS and FIML #57
- p-values #290
- Parallelization of bootstrap #288
- Update to PrettyTables v3 #275
- How about providing z-test and p-value? #215
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
No labels
Activity
[-]v0.4.3[/-][+]v0.5.0[/+]Maximilian-Stefan-Ernst commentedon Jun 21, 2026
@JuliaRegistrator register
Release Notes:
StructuralEquationModels.jl v0.5.0
This is a large, breaking release centered on a modular redesign of the model/type system to unify single models and collections/multigroup models.
Highlights
Semis now a weighted sum of loss terms, each SEM loss bundling its own observed + implied parts.replace_observed.Breaking changes
AbstractLoss/SemLoss;evaluate!is implemented per loss term (nomodelargument) (Refactor objective/gradient/hessian evaluation #214).SemImply→SemImplied(Rename SemImply to SemImplied #235).Semterm structure (Refactor Sem multi-group support #317).SemOptimizerAPI refactor (Refactor SemOptimizer API #299).using NLopt/using ProximalAlgorithmsto enableengine = :NLopt/:Proximal(Support optimization backends via package extensions #228, Switch to ProximalAlgorithms v0.7 #234).params/param_labels(Rename params #253, Rename params (#253) #257);df→dof(rename df => dof #254).χ² = Σ_g (N_g−1)·F_g(new:nsamples_correcteddefault weighting;N−Gχ² multiplier for ML/WLS; matching RMSEA). Multigroup ML/WLS results now match lavaanwishart/GLS; single-group results are unchanged except ML RMSEA. FIML is unchanged.New features
CFIfit measure (Add CFI #313).replace_observedfor collections (Replace observed collections #266).Bug fixes
JuliaRegistrator commentedon Jun 21, 2026
Registration pull request created: JuliaRegistries/General/158625
Tagging
After the above pull request is merged, it is recommended that a tag is created on this repository for the registered package version.
This will be done automatically if the Julia TagBot GitHub Action is installed, or can be done manually through the github interface, or via:
Maximilian-Stefan-Ernst commentedon Jun 21, 2026
@JuliaRegistrator register
Release Notes:
StructuralEquationModels.jl v0.5.0
This is a large, breaking release centered on a modular redesign of the model/type system to unify single models and collections/multigroup models.
Highlights
Semis now a weighted sum of loss terms, each SEM loss bundling its own observed + implied parts.replace_observed.Breaking changes
AbstractLoss/SemLoss;evaluate!is implemented per loss term (nomodelargument) (Refactor objective/gradient/hessian evaluation #214).SemImply→SemImplied(Rename SemImply to SemImplied #235).Semterm structure (Refactor Sem multi-group support #317).SemOptimizerAPI refactor (Refactor SemOptimizer API #299).using NLopt/using ProximalAlgorithmsto enableengine = :NLopt/:Proximal(Support optimization backends via package extensions #228, Switch to ProximalAlgorithms v0.7 #234).params/param_labels(Rename params #253, Rename params (#253) #257);df→dof(rename df => dof #254).χ² = Σ_g (N_g−1)·F_g(new:nsamples_correcteddefault weighting;N−Gχ² multiplier for ML/WLS; matching RMSEA). Multigroup ML/WLS results now match lavaanwishart/GLS; single-group results are unchanged except ML RMSEA. FIML is unchanged.New features
CFIfit measure (Add CFI #313).replace_observedfor collections (Replace observed collections #266).Bug fixes
JuliaRegistrator commentedon Jun 21, 2026
Registration pull request updated: JuliaRegistries/General/158625
Tagging
After the above pull request is merged, it is recommended that a tag is created on this repository for the registered package version.
This will be done automatically if the Julia TagBot GitHub Action is installed, or can be done manually through the github interface, or via: