Mix-Shift and Base-Effect Audit¶
Test / assessment — instantiates Aggregate–Marginal Trajectory Reconciliation
Tests whether composition, denominator, price, seasonality, selection, or comparison base creates apparent divergence.
The Mix-Shift and Base-Effect Audit is a screening test run before any causal or transition story is allowed, asking one question: is the apparent aggregate–marginal divergence real, or is it an artifact of a changing denominator, product or route mix, price level, seasonality, selection, or comparison base? It can close the episode as a false divergence and record the cause. Its defining move is artifact-first suspicion: it tries to explain the divergence away with composition and base effects before anyone interprets it as a genuine transition. What survives the audit is worth governing; what does not is filed as a measurement lesson, not an intervention.
Example¶
An airline's total unit revenue (RASM) is up year over year, while a new set of routes — the entering contribution — shows falling unit revenue, and a strategist is ready to call the new routes a mistake. The audit intervenes. First, the denominator: average stage length on the new routes is much longer, and RASM falls mechanically with distance flown regardless of pricing, so it re-expresses both on a length-adjusted basis. Second, the mix: the new routes serve leisure markets with structurally lower yields, so the "decline" is composition, not deteriorating pricing power. Third, the base: the prior-year comparison quarter carried a fuel surcharge that inflated it. After length, mix, and base adjustments, the sign opposition disappears — the episode closes as a false divergence, with the cause logged so the same illusion is not re-litigated next quarter.
How it works¶
- Freeze the primary definitions first. The frame is fixed before the sign is examined, so the audit cannot be run until it produces a preferred answer.
- Re-express on a common frame. Both trajectories are put on a shared denominator, price basis, and seasonally aligned footing; a divergence that vanishes under alignment was never real.
- Separate composition from within-group performance. The audit distinguishes a shift in who is entering from a change in how a like unit performs.
- Close or hand off. If the opposition survives every adjustment, the case is passed to the transition analysis; if not, it is closed as a false divergence with the cause recorded.
Tuning parameters¶
- Artifacts tested — which of denominator, mix, price, seasonality, selection, and base are checked; skipping one leaves a hole.
- Standardization method — direct versus indirect standardization when holding mix constant; the choice shifts how much "decline" is attributed to composition.
- Season-adjustment approach — how seasonality and calendar effects are removed before comparing direction.
- Selection and representativeness checks — how hard the audit probes whether the entering population changed for a reason that matters.
When it helps, and when it misleads¶
Its strength is that it is the cheapest way to kill a false alarm and the archetype's designated guard against the mix-shift and base-effect masquerade — it stops a composition artifact from being escalated as a transition.
Its failure mode is over-adjustment: standardizing until a real signal is "explained away," or falling into the Simpson's-paradox trap where a mix reversal is dismissed as mere composition when the composition change is the story worth acting on[n1]. The classic misuse is standardizing on a mix that is itself shifting for a meaningful reason. The guarding discipline is to report both the composition (between-group) and within-group effects side by side rather than correcting one away, and to freeze the frame before looking at the sign.
How it implements the components¶
legacy_stock_mix_and_base_effect_decomposition— it isolates composition, price, and base contributions to test whether they alone produce the apparent divergence.aligned_time_denominator_and_vintage_frame— it re-expresses both trajectories on a common denominator, price basis, and seasonally aligned frame before judging the sign.
It tests for composition and base artifacts, not statistical fragility — whether the sign survives the two trajectories' confidence bands and shared-data dependence (uncertainty_and_minimum_sample_guardrail) is the job of its twin Paired Confidence-Band Review; it also does not project the horizon (masking_or_crossover_horizon_estimate — Crossover Scenario Projection).
Related¶
- Instantiates: Aggregate–Marginal Trajectory Reconciliation — the audit is the false-divergence screen that must pass before a transition is declared.
- Consumes: Contribution Waterfall Decomposition supplies the term-by-term bridge the audit inspects.
- Sibling mechanisms: Aggregate–Marginal Sign-Divergence Alert · Cohort or Vintage Analysis · Contribution Waterfall Decomposition · Crossover Scenario Projection · Cumulative-versus-Incremental Dashboard · Dual-Metric Decision Memo · First-Difference or Derivative Estimate · Paired Confidence-Band Review · Rolling Marginal-Contribution Curve
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: Mix-Shift and Base-Effect Audit operates as a bounded evaluation of existing evidence or work that produces a finding or disposition because it tests whether composition, denominator, price, seasonality, selection, or comparison base creates apparent divergence.
Independent corroboration: The frozen evidence defines Mix-Shift and Base-Effect Audit as 'Tests whether composition, denominator, price, seasonality, selection, or comparison base creates apparent divergence', so its operative form is Assessment, Review & Assurance.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Statistics & Experimental Design
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Decomposing apparent change into composition, denominator, seasonality, selection, and comparison-base effects is fundamentally a statistical inference practice.
Related originating lineages:
- Accounting & Auditing — Variance and mix analyses in managerial accounting materially shaped operational versions of the audit.
- Economics & Finance — Index-number analysis, price effects, portfolio composition, and economic base effects independently developed closely related audit conventions.
Review resolution: Both independent reviews agree on primary origin statistics_experimental_design; reconciliation resolves secondary fields (alternate_origin_disagreement, origin_mode_disagreement). Alternate origins retained (accounting_auditing, economics_finance) are the union of reviewer-supported formative lineages with explicit rationales, not a list of later application domains. Present-day breadth is represented separately as domain_reach=multi_domain; origin_mode=cross_disciplinary_synthesis records the historical relationship among lineages. Confidence is conservatively reconciled to high, and encyclopedia_synthesis=false preserves either reviewer's finding that the encyclopedia generalized the mechanism.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] Simpson's paradox: a trend visible within every subgroup can reverse when the subgroups are pooled, because the group sizes — the mix — shift. It is exactly why a mix-and-base audit must report within-group and composition effects separately rather than "correcting away" one of them. ↩