CYDENiC
Ci6 Core · Governed Statistical Processor

Statistics,
computed for AI.

Regression, forecasting, inference, and forensic anomaly detection in one engine — deterministic, closed-form, and reproducible byte-for-byte. No training. No drift. No black box.

The problem

In business-critical work,
99% right can be 100% wrong.

A forecast may tolerate uncertainty. A calculation, test result, compliance determination, or reported number may not.

GSP does not substitute confidence for correctness. It applies validated methods, tests whether the analysis is appropriate, exposes uncertainty, refuses unsupported conclusions, and returns the evidence required to reproduce the result.

Correctly computed. Honestly qualified. Proven on demand.

Anomaly detection
Fraud detection anomalies can't poison.

The tool everyone uses to catch fraud fails because of the fraud. GSP's doesn't.

Input: 50 vendor invoice amounts ($) — 45 clean near $1,000, with 5 fraudulent invoices planted at $2,000. Both screens see the same data.

Naive Detection

mean / standard deviation
0 of 5 frauds caught

The five frauds inflate the standard deviation until their own score falls back under the threshold — the everyday tool hides the very fraud it's hunting.

GSP Robust Detection

median / MAD
5 of 5 frauds caught

A median / MAD band can't be dragged by a few extreme values, so every planted $2,000 invoice lands far outside it and is flagged.

Governance envelope
engineversion-pinned
determinationanomaly screen
inputssha256:d83c32c4…49eab
reproducesbyte-for-byte ✓

Deterministic — the same 50 values reproduce this exact result through the GSP API.

Regression & inference
Regression with the honesty Excel omits.

Excel gives you the coefficient. GSP tells you which ones to believe.

Input: sales ~ marketing_spend + weather_index  (n = 24). One predictor is a real driver; one is noise.
0.994
0.994
Adjusted R²
2.2e-24
Significance F
PredictorCoeftpVerdict
Intercept2.7065.370.000
marketing_spend2.11860.20.000Significant
weather_index0.0130.200.847Not significant

Honesty layer: flags weather_index as an insignificant predictor while marketing_spend (t = 60) is real — fit strength strong. A high R² doesn't make every driver true; GSP names the false one.

Governance envelope
engineversion-pinned
determinationOLS regression
inputssha256:b9fb1c6b…709f54
reproducesbyte-for-byte ✓

Solved by numerically-stable Householder QR — reproducible through the GSP API.

Forensic conformity
Benford's Law — does the ledger's digit structure look natural?

Cooked books leave a fingerprint in the digits. GSP reads it — and tells you when it can't.

Input: 369 ledger amounts spanning many magnitudes — a fabricated set with an unnaturally uniform first-digit spread.
Observed Benford expected Most-deviating digit

Nonconformity · MAD 0.0597. Digit 1 appears 11.1% of the time against a 30.1% Benford expectation — grossly under-represented; digits 8 and 9 over-represented. Flagged for review — a screen, not a finding.

Benford applies only to the right population — thousands of transaction-level, same-unit amounts. The engine discloses when a dataset isn't screenable. That disclosure is the selling point, not the caveat.

Governance envelope
engineversion-pinned
determinationBenford screen
inputssha256:c5e7ee29…584052
reproducesbyte-for-byte ✓

Screened against the published Benford / Nigrini thresholds — reproducible through the GSP API.

Capabilities · soup to nuts

The complete capability set.

Every capability is deterministic and validated against a published reference — and domain-agnostic: the same call runs on a macro series, a P&L, or a raw ledger.

Delivery & integration
For developers

GSP is a governed HTTP API for statistics your app can trust. Send a series and a method — regression, forecast, hypothesis test, fraud screen — and get back the determination, how far to trust it, and a reproducible receipt. Your app renders the pixels; the engine owns the number.

Versioned and SHA-pinnable, so a pinned result reproduces forever. It's HTTP — drop it into any stack, any language. The processors never leave the engine, so you get institutional-grade statistics without building or maintaining a statistics layer of your own.

Pair it with a conversational interface and ship analytics for domains where a wrong number is unacceptable — finance, health, compliance.

Delivery
Governed HTTP API Versioned SHA-pinnable Plot-ready data output Domain-agnostic calls AI-agnostic · BYOK
Anomaly & fraud detection
Flagship

A full forensic battery consolidated into one review queue ranked by cross-method consensus — built on robust estimators the anomalies can't poison. Everyday mean-and-standard-deviation tools inflate their own fence and hide the very fraud they're hunting; these don't. It also discloses when a dataset isn't a valid population to screen, so you never get false confidence — and hands your auditor a defensible, reproducible flag with its reasoning attached.

Forensic battery
Benford's Law · 1-digit Benford's Law · first-two-digit Robust entry screening Grubbs' test Generalized ESD Tukey fences SPC · Shewhart EWMA CUSUM + changepoint Multivariate Mahalanobis Cross-method consensus ranking

Validated against NIST, Rosner, and the published Benford / Nigrini thresholds.

Forecasting

A multi-model out-of-sample bake-off that selects the model which wins on the horizon you actually forecast — never on in-sample fit — then labels the result reliable, directional, or unreliable, or refuses outright when it honestly can't. You get the number and exactly how much to trust it.

Models & outputs
Naive Seasonal Trend Exponential smoothing Model combination Decomposition forecasting Dual predictive bands Scenario projection Out-of-sample selection Reliability verdict
Regression & inference

Full multiple regression with the complete inference output, solved by numerically-stable Householder QR that holds up on ill-conditioned data which breaks naive solvers. It adds the honesty layer spreadsheets omit — insignificant predictors, multicollinearity, and the significant-but-weak trap, named in plain language — so you know which factors truly move the outcome, which don't, and when a "driver" is a false signal.

Outputs & methods
Multiple regression R² / adjusted R² ANOVA F Coefficient t-stats p-values Confidence intervals Residual diagnostics Fit plots Fixed-effects panel models Householder QR solver Multicollinearity detection

Validated against NIST reference datasets — Longley included.

Probability & hypothesis testing

Distributions computed from series and continued fractions, not table lookups, with the classical tests on top. This is the inference foundation that regression p-values, forecast intervals, and conformity tests all stand on — proven against the regression to the last digit: the two-sample t matches the equivalent regression coefficient exactly.

Distributions & tests
Student-t Fisher-F Normal Chi-square One-sample t Two-sample t · Welch Two-sample t · pooled Paired t F-test for variances Goodness-of-fit

Validated against published statistical tables and exact analytic identities.

Time-series & relationships

Relate any series to any other, with the discipline to say "this is context, not cause." Every relationship is labeled association-only — never fabricated causation — significance-gated against multiple-comparison inflation, and reported with its measured strength, so you never take a spurious correlation to your board.

Methods
Trend statistics Lead-lag scanning Seasonal adjustment Threshold classification Cross-sectional divergence Governed context overlays Association-only labeling Multiple-comparison gating
Verify AI-generated claims
The wedge

Point GSP at a claim it can compute — whether a person or an AI asserted it — and it checks that claim deterministically. It turns the engine's identity into a product: the governed, reproducible verifier for the flood of numbers now generated by LLMs. The deterministic fact-checker for the AI era.

Checkable claims
Growth-rate check Driver-effect check Significance check Conformity check
Why it wins

The same job, done honestly.

vs. Spreadsheets

A spreadsheet gives you the number and hides the uncertainty.

GSPgives you the number, the confidence, and the receipt.

vs. LLMs doing math

An LLM generates a confident number that won't reproduce.

GSPdetermines a number that reproduces byte-for-byte.

vs. Black-box ML

A black box can't tell you why.

GSPshows its work — and refuses when it can't.

Honest by construction

It tells you when
not to trust it.

Reliability labels, significance and coverage disclosure, applicability guards, and explicit refusals travel inside every result the engine returns. When the data can't support the question, GSP says so — and refuses an ill-posed input instead of fabricating a confident number.

Reliable Directional Unreliable Refused

The one statistics engine that won't let you fool yourself — or your stakeholders.

Governed determination

Every result ships
with its receipt.

A statistic from the GSP isn't a claim you take on faith. It's a determination — computed by the engine, pinned to versioned logic, and reproducible byte-for-byte with no AI anywhere in the loop. Analytics stop being “trust the model” and start being “here's the proof.”

Computed, not generated
The value comes from the engine's computation, executed the same way every time. The AI operates the experience around it — explaining, orchestrating — but never authors the governed value.
Reproducible
Same inputs, same declared method, same output — byte-for-byte, forever. No training, no random seeds, no drift between one run and the next.
Validated against reference
Checked against NIST reference datasets and the published statistical tables — including the ill-conditioned cases that break naive solvers.
Certifiable
A report-grade result can be sealed as a certified, tamper-evident artifact under a SHA-256 content certificate — a point-in-time record you can file and defend.
Governance envelope — with every result
engineversion-pinned
methodversion-pinned
referenceNIST-validated
inputsSHA-256 hashed
reproducesbyte-for-byte ✓

Hand a regulator, a board, or a court the receipt — and reproduce the number on demand, with no AI in the loop.

Put a statistic you have
to defend on the engine.

Bring a regression, a forecast, or a fraud flag your business runs on — a human's ledger or an AI's claim — and we'll put it on the Governed Statistical Processor and show you the same answer, reproduced and traceable, before anyone talks terms.

By submitting this form, you acknowledge that CYDENiC will use the information provided to review and respond to your request. See our Privacy Policy.

Thank you. We'll be in touch.