Research & Methodology

Methodology

We adhere to a strict "Trust but Verify" doctrine. Synthetic market data is dangerous if it drifts from reality; every run must be deterministic and reproducible.

Calibration Approach (Operator-Driven)

• Venue-specific funding models live in src/aleatoric/venues/* with tests.
• Presets encode starting parameters; adjust manifests for your venue deltas and rerun with fixed seeds.
• Use /mcp/simulate_funding_regime to compare funding paths before committing to a manifest.

Statistical Tests

Suggested checks when comparing against live data:

• KS-Test (Kolmogorov–Smirnov): Verifies that the distribution of synthetic returns matches historical returns.
• Autocorrelation: Ensures volatility clustering mimics real markets.
• Hurst Exponent: Verifies the "memory" of the time series (trending vs. mean-reverting).

Audit & Validation

• Regression coverage: 158 tests, 82% coverage (documented in validation/VALIDATION_REPORT.md).
• Determinism enforced in tests/test_determinism.py.
• Manifest/schema parity in tests/test_mcp_manifest_parity.py.
• Audit logs include job_id, seed, and preset for traceability.