PCM5 Generalized Mixed-Model Evidence Boundary¶
PCM5 extends the cichlid workflow into generalized mixed models and posterior phylogenetic quantities. It is the clearest example of a study family that is fully documented but not yet fully reproduced inside the repository.
flowchart LR
A[Study sources and datasets] --> B[Governed evidence bundles]
B --> C[Documented analytical map]
C --> D[Posterior PGLMM boundary remains open]
What Is Source-Backed And Deployed¶
- the lecture scripts and supplementary code references
- the cichlid observation table, phylogeny, and saved workspace
- governed bundle IDs for each checked-in evidence unit
- the evidence map for posterior signal, ancestral values, count models, infection models, and repeated-measure random effects
What Is Not Yet A Repository-Owned Parity Claim¶
- Poisson phylogenetic mixed-model posteriors
- multinomial mixed-model posteriors
- posterior ancestral-state ledgers
- probability-scale posterior slope transforms
- governed variance partitions matching the lecture's multi-random-effect models
What Readers Can Safely Conclude¶
- the repository now documents the PCM5 lecture family at a much finer-grained evidence-ID level than one generic summary
- the repository preserves the study inputs and analytical decomposition well enough for serious review
- the remaining gap is not source intake or documentation quality; it is the posterior mixed-model reproduction boundary itself
Nearby Repository Support¶
The repository does expose narrower comparative surfaces such as binary-response comparative regression and reviewer-facing model-selection ledgers. Those are useful context, but they are not substitutes for the lecture's generalized mixed-model posterior workflow.
Reading Rule¶
Treat PCM5 as a documented study boundary with a strong source map. Do not read it as proof that the repository already owns a full PGLMM implementation.