compute_assurance() function for unconditional
Bayesian assurance (O’Hagan & Stevens, 2001) computed as a weighted
average of conditional power over a design prior on the effect
size.assurance_prior_weights() convenience wrapper for
constructing normalised design-prior weights (normal, uniform, beta)
over an effect grid.decide_sample_size() function with both assurance
mode (design prior) and conditional mode for recommending sample sizes
from simulation output.validate_inla_vs_brms() function for spot-checking
INLA posterior estimates against brms/Stan.brms_inla_power,
powerbrmsINLA_assurance, and
powerbrmsINLA_sample_size objects.plot_assurance_curve() and
plot_assurance_with_robustness() for unconditional
assurance visualisation.plot_bf_assurance_curve_smooth(),
plot_bf_assurance_curve(),
plot_bf_expected_evidence(), and
plot_bf_heatmap() for Bayes factor visualisation.plot_decision_assurance_curve(),
plot_decision_threshold_contour(), and
add_decision_overlay() for decision-rule
visualisation.plot_design_prior() for visualising design
priors.plot_interaction_surface() for multi-effect grid
visualisation.plot_power_contour(),
plot_power_heatmap(), and
plot_power_assurance_overlay() for conditional power
visualisation.plot_precision_assurance_curve() and
plot_precision_fan_chart().brms_inla_power() now supports multi-effect grids
(data.frame effect_grid), brms-to-INLA prior translation
with full audit trail, marginal-likelihood Bayes factors
(bf_method = "marglik"), and automatic INLA thread
detection.brms_inla_power_sequential() rewritten with
multi-effect support and prior translation.brms_inla_power_two_stage() now uses the modernised
engine internally..to_inla_family(),
.scale_fill_viridis_discrete()).requireNamespace("MASS") guard for negative
binomial data generation..Rbuildignore to exclude
.claude/, .DS_Store, .Rcheck/,
and .tar.gz artefacts.brms_inla_power_parallel() for parallel
simulations.decide_sample_size() and
add_decision_overlay() helpers.error_sd and group_sd now accept
distributional specifications (halfnormal,
lognormal, uniform) for variance-uncertainty
integration; new validate_sd_spec() helper exported.test-validation-classical.R,
test-validation-bayesassurance.R) and accompanying vignette
benchmarking against power.t.test() and
bayesassurance::assurance_nd_na()..github,
LICENSE.md, and cran-comments.md from the
source tarball via .Rbuildignore.
Need a high-speed mirror for your open-source project?
Contact our mirror admin team at info@clientvps.com.
This archive is provided as a free public service to the community.
Proudly supported by infrastructure from VPSPulse , RxServers , BuyNumber , UnitVPS , OffshoreName and secure payment technology by ArionPay.