Prediction Basis Standard (PBAS)
Architecture Ecosystem: Structured Reality Standards™
Architecture Family: PGA™ — Prediction Governance Architecture
Operational Layer: Prediction Basis Governance Layer
Category:
Governance & Enforcement
Subcategory:
Prediction Governance
Type: Parent Standard
Governed Space:
Prediction Basis
Origin ID: OOF-PGA-PBAS
Version: 1.0
Status: Canonical · Open Standard
Effective Date: 7 September 2026
Compatibility: OOF® Methodology OS™ · GOA™ · ORA™ · VALIDOS™ · INTEGROS® · ADIT® · AGA™ · AIG® · ASGA™ · CLIA® · SIMULOS®
AI-Readable: Yes
Authority: OOF®
Protection:
MIP® — Methodological Intellectual Property
Canonical Language:
English (UCL™)
Governed Space
Prediction Basis
Canonical Definition
The Prediction Basis Standard (PBAS) governs the identification, inclusion, exclusion, source relationship, relevance, sufficiency, contextualization, and traceability of the information, data, observations, evidence references, signals, and other materially relevant informational elements that legitimately form the basis upon which a governed prediction is constructed.
PBAS establishes what the prediction is actually based on before PGA™ governs assumptions, prediction formation, probability, uncertainty, feasibility, or predictive performance.
It ensures that the informational foundation of a prediction remains identifiable and traceable rather than being inferred from the final output or reconstructed retrospectively after the outcome becomes known.
Prediction Basis ≠ Evidence Governance
Operational Role
PBAS serves as the third Parent Standard within PGA™ — Prediction Governance Architecture .
It receives:
Governed Prediction Object & Purpose
and:
Governed Prediction Methodology
and establishes the informational foundation upon which that methodology may legitimately operate.
The governed transition is:
Governed Predictive Frame + Governed Prediction Methodology → Governed Prediction Basis
PBAS does not determine whether information is legally admissible evidence, scientifically valid evidence, or sufficient for another governance purpose.
It governs whether and how information participates in the Prediction Basis.
Core Question
What information, data, observations, evidence references, and signals legitimately form the basis of the prediction?
PGA™ Governance Flow
Prediction Object & Purpose → Prediction Methodology → → Prediction Basis → Prediction Assumptions → Prediction Formation → Probability & Uncertainty → Prediction Feasibility → Collective Prediction where
applicable → Communication & Reliance → Monitoring & Revision → Outcome & Performance → Prediction Record & Continuity
Current Position: Prediction Basis
PBAS follows PMES — Prediction Methodology Standard and precedes PAS — Prediction Assumption Standard .
Prediction Basis
A Prediction Basis is the governed set of informational elements upon which Prediction Methodology operates when forming a prediction.
The basis may include:
structured data; unstructured information; measured observations; historical records; real-time signals; model outputs; expert observations; market-derived information; simulation-derived information; external analytical references; evidence references; operational indicators; contextual information; other materially relevant predictive inputs.
PBAS does not assume that every available information source belongs in the Prediction Basis.
Prediction Basis Identity
A governed Prediction Basis must remain identifiable as a distinct informational object associated with a specific prediction or predictive process.
It should be possible to determine:
what information was used,
what information was not used where material,
where it came from,
how it related to the prediction,
and:
which version or state of the basis existed when the prediction was formed.
Prediction Input
A Prediction Input is an informational element that enters or materially informs Prediction Methodology.
Not every available information item becomes a legitimate Prediction Input.
An input becomes governable when its role within the Prediction Basis is sufficiently identifiable.
Available Information ≠ Governed Prediction Input
Prediction Input Identification
PBAS requires the material informational elements forming the Prediction Basis to be identifiable.
Prediction inputs may be:
direct
derived
aggregated
transformed
external
human-generated
machine-generated
or:
hybrid
Their origin does not itself establish their predictive relevance.
Prediction Source
Every material Prediction Input should remain sufficiently associated with its source.
Sources may include:
sensors; databases; public information; private information; human experts; organizations; AI systems; operational systems; prediction markets; research sources; external services; model outputs; simulations; other identifiable source classes.
The source relationship is part of Prediction Basis governance.
Source Mapping
PBAS requires material source relationships to remain traceable.
Source mapping may establish:
Source → Information → Prediction Input → Prediction Basis
where appropriate.
Where information passes through multiple transformations or intermediaries, source lineage should remain identifiable to the degree necessary for prediction governance.
Direct and Indirect Sources
A prediction may rely directly on an original information source or indirectly on information that has already been transformed, summarized, filtered, or interpreted.
PBAS distinguishes:
Original Source
from:
Intermediate Source
and:
Immediate Prediction Input
where those distinctions are material.
Immediate Input Source ≠ Original Information Source
Source Dependence
Multiple Prediction Inputs may originate from the same underlying source.
PBAS must preserve this relationship where it materially affects the informational diversity or apparent independence of the Prediction Basis.
Ten reports derived from one original source do not automatically constitute ten independent informational foundations.
Multiple Inputs ≠ Independent Sources
Information Relevance
Information belongs within Prediction Basis only where its relationship to the governed predictive frame and methodology is sufficiently relevant.
Relevance may concern:
Prediction Object; Predictive Purpose; Prediction Target; Prediction Horizon; Predictive Boundary; selected methodology; methodological requirements; applicable conditions.
PBAS prevents information volume from substituting for predictive relevance.
More Information ≠ Better Prediction Basis
Signal Relevance
A predictive signal is an informational element interpreted as potentially informative about the predicted outcome.
PBAS governs whether a signal is sufficiently relevant to participate in Prediction Basis.
Signal relevance does not establish predictive power.
Signal Presence ≠ Predictive Relevance
Predictive Relevance ≠ Predictive Sufficiency
Signal Strength and Signal Governance
A signal may appear strong while being:
context-specific; historically unstable; dependent on another variable; duplicated across multiple sources; temporally stale; operationally irrelevant; outcome-proximate.
PBAS governs its relationship to Prediction Basis without assuming that signal strength alone establishes legitimacy.
Historical Information
Historical information may form part of Prediction Basis where relevant.
PBAS must distinguish between:
Historical Availability
and:
Current Predictive Relevance
Historical patterns may cease to provide a legitimate basis where structural conditions materially change.
Historical Information ≠ Continuing Predictive Relevance
Current Information
Real-time or current information may materially alter Prediction Basis.
Currentness alone does not establish relevance or sufficiency.
Recent Information ≠ Relevant Information by Default
PBAS governs both temporal freshness and predictive role where material.
Prediction Basis Timing
Prediction Basis must remain temporally anchored to the point at which the prediction is formed.
Information obtained after a prediction was made must not be retrospectively represented as though it formed the original basis.
Later Information ≠ Original Prediction Basis
This principle is essential to historical predictive integrity.
Information Availability
PBAS may govern whether relevant information was:
Available
Unavailable
Partially Available
Delayed
Unknown
at the time of prediction formation.
A missing information element may represent a governance condition rather than an automatic prediction failure.
The absence must remain identifiable where material.
Known Unknowns
Where the prediction is formed despite known missing information, the absence should remain part of the governed Prediction Basis context.
PBAS distinguishes:
Information Not Available
from:
Information Not Considered
and:
Information Excluded
These are not equivalent states.
Information Exclusion
Relevant information may be deliberately excluded for methodological or operational reasons.
Where exclusion is material, the reason should remain governable.
Excluded Information ≠ Nonexistent Information
A prediction may legitimately exclude information, but the exclusion should not disappear from the governance record where it materially affects interpretation.
Information Transformation
Prediction inputs may be transformed before entering methodology.
Examples include:
normalization; aggregation; filtering; cleaning; summarization; feature extraction; embedding; classification; scoring.
PBAS governs the identity and lineage of the informational basis.
PMES governs the methodological logic of the transformation where it forms part of Prediction Methodology.
Derived Information
Derived information is produced from one or more prior information elements.
Its source lineage must remain traceable where material.
Derived Information ≠ Independent Source
This distinction is particularly important where multiple downstream inputs appear independent but originate from the same informational root.
Prediction Basis and Evidence
Information within Prediction Basis may also qualify as evidence under another governance architecture.
PBAS does not determine:
authenticity; evidentiary admissibility; evidentiary sufficiency; evidentiary integrity; audit evidentiary status.
Those questions belong principally to ADIT® and other applicable governance spaces.
PBAS governs a different question:
Did this information legitimately form part of the prediction's informational basis?
Prediction Basis ≠ Evidence Governance
A prediction can legitimately identify the information it used without PGA™ independently certifying that information as evidentially valid.
Likewise, information may be valid evidence for another purpose while remaining irrelevant to a particular prediction.
Valid Evidence ≠ Relevant Prediction Basis by Default
Source Credibility and Prediction Basis
Source credibility may affect whether information should participate in Prediction Basis.
PBAS may preserve source characteristics and relevant limitations.
However, credibility assessment does not automatically become full evidence validation.
Source Reputation ≠ Evidence Validation
Source Reputation ≠ Prediction Relevance
Information Sufficiency
PBAS governs whether the identified informational basis is sufficiently complete for the methodology to proceed within the governed predictive context.
Sufficiency is contextual.
It depends upon:
Prediction Object; Predictive Purpose; Prediction Target; Prediction Horizon; Predictive Boundary; methodology requirements; material information dependencies; material information gaps.
Information Volume ≠ Prediction Basis Sufficiency
Sufficiency Threshold
PBAS does not impose one universal information threshold.
A simple predictive method may require limited information.
A high-consequence or complex predictive process may require a substantially richer Prediction Basis.
The relevant question is whether the basis is sufficient for the methodology and predictive frame being governed.
Incomplete Prediction Basis
A Prediction Basis may be incomplete.
PBAS does not require every prediction with incomplete information to be prohibited.
Instead, the incompleteness must remain visible and governable.
Potential states may include:
Sufficient
Conditionally Sufficient
Incomplete
Materially Deficient
Unable to Determine
Downstream governance may then determine the consequences.
Prediction Basis Gaps
A Prediction Basis Gap exists where material information required for methodological or predictive governance is missing, unavailable, excluded without sufficient justification, untraceable, or insufficiently relevant.
PBAS treats gaps as governance conditions requiring clarification rather than automatically assuming misconduct or invalidity.
Conflicting Information
Prediction Basis may contain conflicting signals or information.
Conflict must not be silently removed merely to create a coherent narrative.
Where material, PBAS requires the existence of informational conflict to remain identifiable.
Resolution logic may belong to Prediction Methodology where the methodology specifies how conflicting inputs are handled.
Conflicting Information ≠ Invalid Prediction Basis by Default
Redundant Information
Multiple information items may duplicate the same underlying signal.
PBAS must prevent redundancy from being mistaken for independent informational confirmation.
Repeated Signal ≠ Independent Confirmation
Information Diversity
Information diversity concerns materially different informational origins, perspectives, measurements, or mechanisms.
Diversity may strengthen a Prediction Basis in some contexts but does not automatically establish quality or sufficiency.
Information Diversity ≠ Prediction Basis Sufficiency
Predictor–Reality Proximity
PBAS is the primary PGA™ Parent Standard governing the informational relationship between predictive sources and Operational Reality.
Predictor–Reality Proximity represents the governed relationship between a predictor or predictive source and the information, actors, processes, authorities, systems, decisions, or operational conditions from which the predicted outcome may emerge or by which that outcome may be materially determined or influenced.
Potential proximity dimensions may include:
information access; information timing; information exclusivity; organizational proximity; decision proximity; operational proximity; authority proximity; non-public information exposure; outcome knowledge; influence capability; control capability.
Predictor–Reality Proximity Continuum
Where applicable, predictive sources may be situated along a continuum such as:
Public-Information Predictor → Information-Advantaged Predictor → Decision-Proximate Predictor → Outcome- Informed Predictor → Outcome-Influencing Predictor → Outcome-Controlling Predictor
The continuum is descriptive governance structure.
It does not establish misconduct.
High Predictor–Reality Proximity ≠ Misconduct
Information Advantage
A predictor may have access to information unavailable to others.
This may materially affect predictive performance.
PBAS requires the informational condition to remain distinguishable from methodological predictive capability.
Information Advantage ≠ Predictive Capability
A predictor may be highly accurate because it possesses superior information rather than superior forecasting methodology.
Outcome Knowledge
A source may possess knowledge of an outcome before the outcome becomes public or observable to other predictors.
This condition is materially different from forecasting the outcome.
Outcome Knowledge ≠ Predictive Capability
Knowing an outcome before others is not the same as predicting it better than others.
PBAS preserves the informational distinction without determining misconduct.
Decision Proximity
A predictor may be close to people or systems capable of determining or materially affecting the outcome.
Decision proximity must remain distinguishable from decision authority.
Decision Proximity ≠ Decision Authority
Likewise:
Outcome Knowledge ≠ Outcome Control
Outcome Influence
Where a predictive source can materially influence the outcome being predicted, the informational relationship may become relevant to later prediction communication, reliance, monitoring, and performance interpretation.
PBAS identifies relevant source relationships.
PCRS and PMRS govern later prediction influence and changed predictive conditions where applicable.
Non-Public Information
Non-public information may form part of Prediction Basis where legally and methodologically permissible.
PBAS does not determine legal authorization to possess or use that information.
It governs whether such information formed part of the Prediction Basis and whether its role remains identifiable.
Non-Public Information ≠ Misconduct by Default
Legal and compliance determinations remain outside PGA™.
Source Conflict of Interest
A predictive source may have interests related to the predicted outcome.
Where this relationship materially affects Prediction Basis governance, it should remain identifiable.
PBAS does not allocate accountability or determine wrongdoing.
Potential Conflict ≠ Established Misconduct
AI-Generated Information
AI systems may produce information that becomes part of Prediction Basis.
PBAS must preserve whether the informational element was:
directly observed; retrieved; generated; inferred; synthesized; transformed.
AI-Generated Information ≠ Observed Operational Reality
AI governance remains with the relevant OOF® architecture.
Simulation-Derived Information
Simulation outputs may participate in Prediction Basis.
PBAS must preserve their origin as simulation-derived.
Simulation Output ≠ Operational Observation
Simulation-Derived Information ≠ Observed Reality
SIMULOS® retains governance ownership of simulation.
Market-Derived Information
Market prices, prediction market prices, implied probabilities, trading activity, or market sentiment may form part of Prediction Basis.
PBAS preserves their origin and source relationships.
It does not assume:
Market Price = Probability
or:
Market Signal = Population Belief
Probability representation and collective prediction governance belong to later PGA™ Parent Standards.
Probability Provenance Relationship
Where a probability-like output already exists within Prediction Basis, its origin should remain identifiable.
Potential origins include:
statistical estimate; model output; expert judgment; AI forecast; prediction market; ensemble; simulation-supported estimate; hybrid method.
Same Probability Value ≠ Same Predictive Meaning
PMES governs methodology provenance.
PBAS governs the informational source relationship where that output participates as Prediction Basis.
PPUS later governs probability and uncertainty representation.
Prediction Basis Change
Prediction Basis may change as new information becomes available.
A changed basis does not automatically rewrite the basis of an earlier prediction.
Current Prediction Basis ≠ Historical Prediction Basis
PBAS therefore preserves basis state relative to prediction formation time.
Prediction Basis Decay
A Prediction Basis may lose predictive relevance over time.
This may occur through:
stale information; structural change; expired observations; changed operational conditions; changed availability; superseded data; emerging contradictory signals.
Valid Initial Prediction Basis ≠ Valid Continuing Prediction Basis
PMRS later governs whether such change triggers prediction review or revision.
Basis Continuity
Prediction Basis Continuity requires that the informational foundation of the prediction remain reconstructable across later lifecycle stages.
It should be possible to determine:
what was known,
what was used,
what was unavailable,
and:
what changed later.
Retrospective Basis Drift
PBAS prevents later information from being silently incorporated into the interpretation of an earlier prediction.
An analyst reviewing a successful prediction after the outcome becomes known must not assume that information available later formed part of the original basis.
Later Knowledge ≠ Earlier Prediction Basis
Prediction Basis Sufficiency Determination
A governed basis should support an explicit sufficiency determination.
Possible outcomes may include:
Sufficient
Conditionally Sufficient
Clarification Required
Insufficient
Unable to Determine
The determination governs the informational basis only.
It does not establish prediction validity or correctness.
Governance Requirements
PBAS requires that:
1. material Prediction Inputs remain identifiable; 2. material sources remain traceable; 3. source dependence remain visible where relevant; 4. Prediction Inputs be distinguishable from merely available information; 5. material signal relevance be established; 6. historical and current information remain distinguishable; 7. information available after prediction formation not be retrospectively treated as original Prediction Basis; 8. material exclusions and unavailable information remain identifiable; 9. derived information preserve relevant source lineage; 10. information redundancy not be mistaken for independent confirmation; 11. material information conflicts remain visible; 12. Prediction Basis sufficiency be explicitly governable; 13. Predictor–Reality Proximity be preserved where materially relevant; 14. information advantage remain distinguishable from predictive capability; 15. outcome knowledge remain distinguishable from prediction; 16. simulation-derived, AI-generated, market-derived, and operationally observed information remain distinguishable; 17. Prediction Basis changes remain historically traceable; 18. Prediction Basis Decay be capable of triggering downstream review; 19. current basis not overwrite historical basis; 20. complete basis traceability remain possible across the Prediction Lifecycle.
Governance Logic
PBAS applies the following governance logic:
Governed Predictive Frame + Governed Prediction Methodology
↓
Prediction Inputs Identified
↓
Sources Identified and Mapped
↓
Source Dependence Assessed
↓
Signal / Information Relevance Assessed
↓
Included / Excluded / Unavailable Information Identified
↓
Derived Information Lineage Established
↓
Material Information Conflicts / Redundancies Identified
↓
Predictor–Reality Proximity Assessed where applicable
↓
Information Advantage / Outcome Knowledge / Influence Relationships Preserved where material
↓
Prediction Basis Sufficiency Determined
↓
Governed Prediction Basis
↓
PAS — Prediction Assumption Standard
Where sufficiency cannot be established:
Prediction Basis Governance Gap → Clarification / Additional Information / Restriction / Reassessment
Governance Outputs
PBAS may produce:
Governed Prediction Basis; Prediction Input Register; Prediction Source Map; source-dependency relationships; signal-relevance determination; included-information record; excluded-information record;
unavailable-information record; derived-information lineage; information-conflict record; redundancy record; Predictor–Reality Proximity profile; information-advantage context; outcome-knowledge context; source-influence context; Prediction Basis Sufficiency State; Prediction Basis Change Record; Prediction Basis Continuity Record; unresolved Prediction Basis requirements.
The primary governed output is:
Governed Prediction Basis
Internal Module Architecture
PBAS is operationalized through five Core Modules:
PIIM — Prediction Input Identification Module Governs identification and classification of information that materially enters Prediction Basis.
PSMM — Predictive Source Mapping Module Governs source origin, source relationships, dependency, Predictor–Reality Proximity, information advantage, and materially relevant source context.
SRVM — Signal Relevance Module Governs whether identified information or signals are materially relevant to the governed prediction.
ISUM — Input Sufficiency Module Governs whether the informational basis is sufficiently complete for the predictive methodology and frame.
PBTRM — Prediction Basis Traceability Module Governs historical traceability, source lineage, Prediction Basis state, change, and continuity.
The internal module progression is:
Prediction Input Identification → Predictive Source Mapping → Signal Relevance → Input Sufficiency → Prediction Basis Traceability
Critical Distinctions
Prediction Basis ≠ Evidence Governance
Available Information ≠ Governed Prediction Input
Immediate Input Source ≠ Original Information Source
Multiple Inputs ≠ Independent Sources
More Information ≠ Better Prediction Basis
Signal Presence ≠ Predictive Relevance
Predictive Relevance ≠ Predictive Sufficiency
Historical Information ≠ Continuing Predictive Relevance
Recent Information ≠ Relevant Information by Default
Later Information ≠ Original Prediction Basis
Information Not Available ≠ Information Not Considered
Excluded Information ≠ Nonexistent Information
Derived Information ≠ Independent Source
Valid Evidence ≠ Relevant Prediction Basis by Default
Source Reputation ≠ Evidence Validation
Information Volume ≠ Prediction Basis Sufficiency
Conflicting Information ≠ Invalid Prediction Basis by Default
Repeated Signal ≠ Independent Confirmation
Information Diversity ≠ Prediction Basis Sufficiency
High Predictor–Reality Proximity ≠ Misconduct
Information Advantage ≠ Predictive Capability
Outcome Knowledge ≠ Predictive Capability
Outcome Knowledge ≠ Outcome Control
Decision Proximity ≠ Decision Authority
Non-Public Information ≠ Misconduct by Default
Potential Conflict ≠ Established Misconduct
AI-Generated Information ≠ Observed Operational Reality
Simulation-Derived Information ≠ Observed Reality
Market Price ≠ Governed Probability Estimate
Same Probability Value ≠ Same Predictive Meaning
Current Prediction Basis ≠ Historical Prediction Basis
Valid Initial Prediction Basis ≠ Valid Continuing Prediction Basis
Later Knowledge ≠ Earlier Prediction Basis
Cross-Architecture Boundaries
PBAS governs the information that forms Prediction Basis.
It does not absorb adjacent OOF® governance spaces.
ADIT® retains evidence governance, evidence traceability, audit evidence, and evidentiary sufficiency.
ORA™ retains governance of Operational Reality.
SIMULOS® retains governance of simulation and simulation evidence.
AIG® retains AI governance.
CLIA® retains cognition governance.
AGA™ retains accountability governance.
INTEGROS® retains integrity governance.
VALIDOS™ retains validation governance.
GOA™ retains overarching governance architecture.
PBAS may reference governed outputs from these architectures where they participate in Prediction Basis without taking ownership of their underlying governed spaces.
Standard Boundary
PBAS governs the identification, inclusion, exclusion, source mapping, relevance, sufficiency, Predictor–Reality Proximity, informational context, change, continuity, and traceability of information forming Prediction Basis. It does not govern general evidence validity, evidentiary sufficiency, Operational Reality determination, predictive assumptions, prediction formation, probability representation, uncertainty, feasibility, collective prediction legitimacy, communication and reliance, predictive performance, accountability, or final prediction validity.
PBAS governs:
what the prediction is based on.
It does not determine:
whether the prediction is correct.
Canonical Closing Statement
A prediction cannot be meaningfully governed if the information beneath it cannot be identified, traced, and distinguished from what became known only later.
PBAS establishes the governed informational foundation upon which the prediction may legitimately proceed.
Nahlad
Prediction Basis Standard
Architecture:
PGA™ — Prediction Governance Architecture
Category:
Governance & Enforcement
Subcategory:
Prediction Governance
Canonical Definition: The Prediction Basis Standard governs the information, data, observations, evidence references, and signals that legitimately form the informational foundation of a prediction.
Governed Space:
Prediction Basis
Core Question: What information legitimately forms the basis of the prediction?
Prediction Basis ≠ Evidence Governance
→ View Standard
About standard
Governed Links
About Prediction Basis Standard (PBAS)
Prediction Information Identification Module (PIIM)
Prediction Signal Mapping Module (PSMM)
Source Relevance & Validity Module (SRVM)
Information Sufficiency Module (ISUM)
Prediction Basis Traceability Module (PBTRM)