Loading
Disclosure
Calendar and celestial overlays are behavioural finance inputs that track actor beliefs, not independent causal signals. They carry no convergence weight and the system performs comparably without them.
A multi-layer intelligence system that monitors four primary signal layers plus a narrative overlay, scores their convergence, and synthesises high-conviction intelligence briefs. Every assessment traces back to observable data. Every prediction is tracked, timestamped, and scored against outcomes.
Don't take our word for it. View the live prediction record with timestamped forecasts, stated probabilities, resolution outcomes, and Brier scores computed in real time.
Each primary signal layer operates on fundamentally different data sources with different dynamics. This independence is what makes convergence meaningful. Calendar and celestial data serve as narrative context for understanding actor beliefs, not as independent predictive signals.
NEXUS does not predict events. It identifies conditions under which events become more probable. The distinction matters: prediction implies certainty, while convergence analysis surfaces elevated probability windows.
Every output feeds back into the system. Weights, thresholds, and scoring parameters are living values that evolve with each completed prediction cycle. The system never assumes its current calibration is final.
Every assessment traces back to specific, observable data points. The system does not hallucinate connections or project patterns that aren't supported by the underlying signals.
Each phase builds on the previous. Click to expand.
More primary layers converging means a stronger signal. Two-layer convergences are common. Three or more are significant. Full four-layer convergence is exceptionally rare and always flagged as critical. Narrative overlays (CAL/CEL) add context but do not contribute convergence weight.
Signals that cluster tightly in time score higher than the same signals spread over weeks. The scoring function applies a time-decay weighting that favours narrow convergence windows.
Every convergence event is matched against a library of past convergences. Patterns with strong historical precedent and clear outcome data receive higher confidence scores.
Convergence scoring is non-linear and only counts primary signal layers (GEO, MKT, OSI, and additional data layers). Narrative overlays (CAL/CEL) provide actor-belief context but do not contribute convergence weight. The amplification curve steepens as more primary layers align, reflecting the decreasing probability of coincidental overlap.
The system follows a three-step pipeline: Bayesian signal fusion produces a calibrated posterior, the posterior anchors forecast generation, and forecasts are scored against outcomes with a proper scoring rule. The live prediction record shows every forecast, its stated probability, outcome, and score contribution.
Each signal layer contributes evidence to a Bayesian fusion engine that produces a calibrated posterior probability. The system operates in log-odds space to guarantee valid probability outputs regardless of input magnitude.
Each layer contributes a likelihood ratio weighted by its historical reliability. Low-significance signals produce modest shifts. High-significance signals from reliable layers move the posterior substantially. Reliability weights are recalibrated continuously against resolved predictions.
A proprietary discount mechanism prevents double-counting when signal layers carry correlated evidence. The system distinguishes genuinely independent signals from those that share upstream causes.
Only primary layers (GEO, MKT, OSI, SYS) contribute convergence weight. The amplification curve is non-linear, reflecting the decreasing probability of coincidental overlap as more independent layers align. Narrative overlays contribute zero convergence weight.
Calendar and celestial signals are actor-belief context. They capture what market participants think matters. The narrative contribution is hard-capped at a deliberately small level. Removing it entirely does not materially change system accuracy.
This is the step most systems hide. The posterior probability from Step 1 is the quantitative anchor, but the final forecast confidence incorporates regime context, game-theoretic structure, and historical precedent. This function includes structured analytic tradecraft that resists reduction to algebra. We state this explicitly rather than hiding it behind an arrow.
The forecast confidence is tightly bounded around the quantitative posterior. The AI does not override the signal; it provides contextual adjustment within strict limits. Every forecast is scored against actual outcomes, so any systematic bias is caught and penalised.
The signal fusion output anchors the forecast. The AI cannot deviate beyond strict bounds.
Volatility regime, risk appetite, and commodity conditions shape how the posterior is interpreted. Wartime classification invalidates active predictions.
Incomplete-information game modelling identifies conditions under which conflict or cooperation becomes structurally likely.
A proprietary intelligence store of historical precedents, resolved predictions, and analyst notes retrieved via semantic similarity.
Every forecast is scored against actual outcomes using a proper scoring rule with time-decay weighting. Recent predictions count more than older ones. This is the mechanism that keeps the system honest: the only way to optimise the score is to state your true beliefs. Overconfidence is penalised. Calibration is rewarded.
Predictions are resolved as confirmed, partial, or denied. Each resolution feeds back into every upstream component to recalibrate weights, thresholds, and detection parameters.
The scoring rule is strictly proper: it cannot be gamed by hedging, and it penalises both overconfidence and underconfidence symmetrically. Every score is published on the live prediction record.
Scores decompose into calibration error, confidence scatter, and discrimination power. Positive discrimination means the signal framework has genuine predictive value beyond chance.
The system operates on a set of calibrated constants: layer reliability weights, a sensitivity parameter, convergence amplifiers, narrative caps, scenario base rates, and a decay half-life. Initial values were set from domain knowledge and calibrated against the first cohort of resolved predictions.
These are living values: as the prediction record grows, they are recalibrated against out-of-sample outcomes. The specific constants are proprietary. Current performance is visible on the live prediction record. Signal decay models disconfirmation as reversion to prior rather than negative evidence, a deliberate design choice that bounds the likelihood ratio for the current signal space.
Specific data providers, API configurations, and ingestion pipelines are proprietary. The categories above describe the types of data consumed, not the specific sources or methods used to acquire them.
Truly unprecedented events have no historical pattern to match against. The system can detect unusual conditions but cannot anticipate events with no precedent.
Some signal layers operate on delayed data. Geopolitical and OSINT signals may lag minutes to hours behind real-time events. Market data varies by feed tier.
Market regimes change. Correlations that held during one period may break down in the next. The feedback loop mitigates this but cannot eliminate it entirely.
The prediction record is still young. Brier scores and calibration curves become more reliable with hundreds of resolved predictions. We publish confidence intervals on all metrics and flag when sample sizes are insufficient for reliable conclusions.
Deep dive into signal detection, intensity scoring, decay functions, and cross-layer amplification.
Read moreNarrative context layers: historical calendar-market patterns as actor-belief overlay, not independent signals.
Read moreLive accuracy tracking, Brier scores, and performance breakdowns by signal layer and time horizon.
Read moreNash equilibria, Schelling focal points, and escalation ladders applied to geopolitical scenario modelling.
Read moreAccess the full NEXUS platform to explore live signal detection, convergence analysis, and AI-driven intelligence briefs.
Request Access