Advanced Optional Bayesian Workflows¶
Python-facing port of
advanced-optional-workflows.Rmdfrom the frozen R gp3bayes 0.5.0 reference. The statistical and governance framing below follows the canonical vignette; executable Python workflows use the mapped APIs listed later.
This article describes the optional post-0.1.1 extensions. They remain contract-first: neither unrestricted formulas nor automatic model selection are introduced.
Capability audit¶
Separate interaction priors¶
The binary advanced default is normal(0, 0.75) for population main effects
and normal(0, 0.50) for the single approved interaction. The duration
advanced defaults are 0.35 and 0.25 respectively. These are candidate workflow
defaults and still require prior-predictive review.
Full-MCMC backend selection¶
The advanced fitting functions accept only rstan or cmdstanr, and they
always use full MCMC sampling.
Separation screening¶
The screen is a fixed-effects design diagnostic. It is not a replacement for the hierarchical Bayesian fit or its posterior diagnostics.
PSIS-LOO and model averaging¶
The comparison reports predictive differences and diagnostics but never selects a model automatically.
Power-scaling sensitivity¶
Low local sensitivity is not a proof of universal robustness.
Simulation-based calibration¶
The brms generator and brms inference backend share implementation code. An independently coded generator is preferable when the goal is to identify shared implementation defects.
Python API mapping¶
gp3bayespy.assess_powerscaled_sensitivitygp3bayespy.bayesian_backend_capabilitiesgp3bayespy.compare_psis_loogp3bayespy.compute_loo_model_weightsgp3bayespy.compute_psis_loogp3bayespy.create_brms_sbc_plangp3bayespy.create_model_contractgp3bayespy.detect_binary_separationgp3bayespy.fit_binary_model_backendgp3bayespy.interaction_prior_summarygp3bayespy.prepare_hierarchical_binary_datagp3bayespy.run_sbc_plangp3bayespy.simulate_hierarchical_binary_datagp3bayespy.specify_binary_model_with_interaction_prior
Python usage¶
import gp3bayespy as gp
# All functions listed above are available from the package root.
# Use help(gp.<function>) or the API reference for the exact Python signature.