/modelling/index.html Purpose Describe model classes, inference contracts, diagnostics, and decision-layer semantics for dsambayesplus. This section is primarily reference material; use Principled Bayesian Workflow for the methodology spine. Audience Practitioners building and interpreting dsambayesplus models. Reviewers validating modelling assumptions and outputs. Pages Page Topic Estimator Capabilities Approved RE, CRE, and FE scope, implementation status, rejected combinations, and qualification gates Model Classes BLM, hierarchical, and pooled class constructors, fit support, and limitations Model Object Lifecycle State transitions from construction through fitting to post-fit extraction Priors and Boundaries Prior schema, defaults, overrides, boundary controls, and scale semantics Minimal-Prior Policy Governance guidance for prior specification in MMM Response Scale Semantics Identity vs log response, KPI-scale conversion, Jensen-safe reporting Diagnostics Gates Policy modes, threshold tables, identifiability gate, and remediation actions CRE / Mundlak Correlated random effects for hierarchical models Time Components Managed holiday feature generation and weekly anchoring Counterfactual Response Posterior fitted-response paths, draw-wise horizon totals, lift and uncertainty Budget Optimisation Decision-layer budget allocation, objectives, risk scoring, and response transforms Hugo en-gb /modelling/model-classes/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/model-classes/index.html Purpose dsambayesplus provides four model classes for Bayesian marketing mix modelling. Each class targets a different data structure and pooling strategy. This page describes the constructor pathways, fit support, and practical limitations of each class so that an operator can select the appropriate model for a given dataset. The fixed-effects (FE) class is a separate within-unit estimator. A hierarchical model with unit random intercepts remains a random-effects model, not an FE estimator. /modelling/model-object-lifecycle/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/model-object-lifecycle/index.html dsambayesplus model objects (blm, hierarchical, pooled) are mutable S3 lists that progress through a well-defined sequence of states. Understanding these states helps avoid calling post-fit accessors on an unfitted object or forgetting to compile before fitting. This page is an API/runtime reference. It explains how dsambayesplus model objects move through construction, compilation, fitting, and post-fit access. It is not the main guide for prior-setting, diagnostics meaning, or model adequacy. For that, use the Principled Bayesian Workflow. /modelling/priors-and-boundaries/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/priors-and-boundaries/index.html For the workflow guidance behind these controls, start with Stage 2: Model and Priors. This page is the technical contract for dsambayesplus prior and boundary behaviour. Use this page when you need exact dsambayesplus semantics: supported prior families, override syntax, default generation rules, and scaling behaviour. Do not use it as the main argument for why a prior is reasonable. That reasoning belongs in the workflow pages and in your modelling rationale. /modelling/minimal-prior-policy/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/minimal-prior-policy/index.html This page is the short operating rule for prior-setting in dsambayesplus. Use it when you want a compact default policy. For the full workflow logic, see Stage 2: Model and Priors. For the mechanics of YAML and API prior specification, see Priors and Boundaries. Purpose Use a principled but low-friction prior setup that avoids specification-hunting while preserving identifiability in short, collinear MMM datasets. /modelling/response-scale-semantics/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/response-scale-semantics/index.html Purpose dsambayesplus models can operate on an identity (level) or log response scale. This page defines how response scale is detected, stored, and used for post-fit reporting, so that operators understand which scale their outputs are on and how KPI-scale conversions work. Response scale detection Response scale is determined at construction time by detect_response_scale(), which inspects the left-hand side of the formula: /modelling/cre-mundlak/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/cre-mundlak/index.html Purpose The correlated random effects (CRE) pathway, implemented as a Mundlak device, augments hierarchical dsambayesplus models with group-mean terms. This separates within-group variation from between-group variation for selected regressors. It addresses correlation between group effects and regressors when the conditional mean of the selected random intercept is adequately represented by the included group means. It does not correct correlated random slopes, time-varying endogeneity, or omitted time-varying confounders. /modelling/time-components/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/time-components/index.html Purpose dsambayesplus provides managed time-component generation through the effects.holidays config section. When enabled, the runner deterministically generates holiday feature columns from a calendar file and appends them to the compiled model formula. This page defines the configuration contract, generation logic, naming conventions, and audit properties. Overview Time components in dsambayesplus cover: /modelling/diagnostics-gates/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/diagnostics-gates/index.html For the workflow interpretation of these checks, start with Stage 4: Computation and Sampler and Stage 5: Model Adequacy. This page is the threshold and policy reference. Use this page when you need exact gate thresholds, status aggregation, or YAML policy semantics. It does not replace substantive model review: a model can clear threshold tables and still be a poor basis for interpretation. Geographic FE/RE/CRE models also have separate pre-fit panel design diagnostics, including same-date geographic contrasts and conditional design support. /modelling/estimator-capabilities/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/estimator-capabilities/index.html Current implementation status This document is the approved Gate A contract for hardening the random-effects (RE), correlated-random-effects (CRE), and fixed-effects (FE) estimators. Approval fixes the intended scope. It does not mean that every target capability is implemented or production qualified. The current implementation status is authoritative until the relevant implementation and evidence gates have passed: Estimator Current status Safe interpretation RE supported_with_limits Existing Gaussian hierarchical estimator. Its population slopes require conditional mean independence between group effects and regressors. Retained block-support diagnostics are available; Monte Carlo qualification is pending. CRE supported_with_limits Enforced random-intercept Mundlak augmentation for simple externally transformed or untransformed numeric regressors. Retained between-design and hierarchy block-support diagnostics are available; Monte Carlo qualification is pending. FE implemented_not_qualified The direct R API and bounded YAML runner implement Gaussian FE with exact within-unit contrasts. The direct API reconstructs conditional retained-unit intercepts and level output after MCMC. P4-J technical gates passed, but deterministic one-replication checks do not provide repeated-sampling qualification. No estimator is production qualified by this document alone. Production status requires the implementation, package validation, Stan review where applicable, and predeclared Monte Carlo evidence. /modelling/counterfactual-response/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/counterfactual-response/index.html Purpose counterfactual_response() evaluates one fitted MCMC model under two aligned predictor paths. It returns posterior draws of the fitted response location under the proposed scenario, the reference response, and their difference. The result is a model-implied scenario contrast. It is not, by itself, a causal effect or an attribution estimate. Causal interpretation requires a credible identification argument for the fitted model and the intervention being represented. /modelling/budget-optimisation/index.html Mon, 01 Jan 0001 00:00:00 +0000 /modelling/budget-optimisation/index.html Purpose dsambayesplus provides a decision-layer budget optimisation engine that operates on fitted model posteriors. Given a channel scenario with spend bounds, response-transform specifications, and an objective function, the engine searches for the allocation that maximises the chosen objective while respecting channel-level constraints. This page defines the inputs, objectives, risk scoring, response-scale handling, and output structure. Overview Budget optimisation is separate from parameter estimation. It takes a fitted model and a scenario specification, then: