Hidden Patterns in Plain Sight
Zoltar's Sign-Invariant Cascading Threshold Classifier conditions on 27 distinct market regimes across 150+ years of data — surfacing interpretable monthly directional predictions with Bayesian confidence scores.
A true mathematical innovation at the core. One unified monthly report spanning US equities (the S&P 500) and select commodities — gold, silver, platinum, and oil.
A True Mathematical Innovation
Zoltar owns its intellectual property — a regime-conditional, sign-invariant framework for fusing disparate signals. Structured, data-efficient, and transparent by design.
Regime-Aware
Each signal is judged only in the market states where it actually holds — never blended into a single average that fits nothing.
Sign-Invariant by Design
A reliably-wrong signal carries the same information as a reliably-right one. Zoltar corrects for it automatically — no manual intervention.
Walk-Forward Valid
Every feature uses only trailing data. Nothing peeks at the future — the backtest is evaluated honestly out-of-sample.
Disciplined
When the models disagree, Zoltar abstains rather than forcing a call. Most months, the conviction bar isn't met — and that restraint is the point.
Fully Interpretable
No black box. Every call traces to named, inspectable rules — transparent where deep nets are opaque.
Whitepaper-Backed
The framework is formalized — with proofs and walk-forward validation — in a public whitepaper you can read in full.
How Zoltar Works
The SICTC engine — regime-conditional, sign-invariant, and interpretable by design.
Condition on the Regime
REGIME-AWARE
Every month is first classified by its market backdrop — because a signal that works in one regime can mislead in another.
Sign-Invariant Cascade
THE CORE IDEA
A cascade of models judges each signal only where it applies, and corrects reliably-wrong signals automatically. This is the heart of the SICTC method.
Monthly Call — or Abstain
YOUR REPORT
Where independent models converge, you get a high-conviction monthly directional call with a confidence score. When they disagree, Zoltar abstains rather than force a call.
The full method — the regime partition, threshold cascade, and quality scoring — is derived, with proofs and validation, in the whitepaper.
Download the WhitepaperWhat You Receive
Every month, our pipeline processes the latest data and delivers a comprehensive report covering US equities and select commodities. Every signal is transparent — you see the confidence, the historical basis, and the statistical rigor behind it.
Our strength lies in convergence: when CAPE valuation, macro momentum, liquidity regime, and cross-asset signals all agree, conviction is highest.
How We Compare
Zoltar Predicts delivers what no single alternative can: quantitative rigor, historical depth, and multi-model convergence — all in one monthly report.
| Feature | Zoltar |
|---|---|
| 150+ Years Historical Data | |
| Bayesian Confidence Scores | |
| Regime-Conditional Models | |
| Interpretable Methodology | |
| Multi-Asset Coverage | |
| Walk-Forward Backtests | |
| Monthly Structured Reports | |
| Published, Formal Methodology |
Ready to See the Signals?
Start with Macro Monthly — your first three months are free. Directional predictions, confidence scores, and the full report, every month. Financial markets are just the beginning.
$50 for a 30-minute consultation. No obligation.
Investment Risk Disclosure
Zoltar Predicts is not a registered investment advisor, broker-dealer, or financial planner. All content, analysis, predictions, and reports provided through this platform are for informational and educational purposes only and do not constitute financial advice, investment advice, or a recommendation to buy, sell, or hold any securities.
Past performance of our models does not guarantee future results. All investments carry risk, including the potential loss of principal. You should conduct your own research and consult with a licensed financial advisor before making any investment decisions. By using our services, you acknowledge and accept these risks.
Any performance figures shown are backtested or hypothetical, have inherent limitations, and do not represent actual trading results or a live track record.