Deploy Model
Zorynex data terminal interface showing portfolio analysis charts
Backtested Decision Support

Predictive Portfolio Analysis, Backtested Against More Than a Decade of Market Data

Zorynex applies statistical models to historical and live market data to flag risk, surface rebalancing opportunities, and quantify the trade-offs behind each recommendation before you commit capital.

View Backtest Results How the Model Works
ZRX-MODEL OUTPUT● SIMULATION
Illustrative backtest curve — not a forecast and not indicative of future returns.
Servers located within the EU
Model logic logged for every output
No automated execution without confirmation
Built for individual, not institutional, account sizes
Core Capabilities

What the Model Actually Computes

Each module below runs continuously against your portfolio configuration rather than producing a one-off report.

Predictive Modeling

Pattern Recognition on Historical Series

The model is trained on historical price, volume, and volatility series to estimate the probability of specific market conditions recurring. Output is a ranked set of scenarios, not a single guaranteed number.

Risk Mitigation

Backtested Volatility Bands

Positions are evaluated against drawdown thresholds you set in advance. When backtested volatility exceeds that threshold, the position is flagged for review before any allocation changes.

Real-Time Processing

Continuous Data Ingestion

Market data pipelines refresh continuously rather than on a daily batch cycle, so model output reflects intraday conditions rather than yesterday's close.

Scalability

From One Portfolio to Several

The same engine runs whether you track a single supplemental portfolio or multiple accounts with separate risk profiles, without requiring separate infrastructure.

Methodology

Inside the Backtesting Engine

Alpha generation claims are only useful if the underlying test is transparent. Here is the sequence the engine actually runs.

  1. 01

    Data Ingestion

    Historical price series, corporate actions, and macro indicators are normalized into a single time-aligned dataset.

  2. 02

    Feature Engineering

    Volatility, momentum, and correlation features are derived from the raw series and weighted by relevance to the chosen strategy.

  3. 03

    Strategy Simulation

    The candidate strategy is run against rolling historical windows to produce a distribution of outcomes, not a single result.

  4. 04

    Statistical Validation

    Results are checked for statistically significant deviation from random outcomes before a strategy is surfaced as viable.

Why Backtest Length Matters

A strategy tested over a single market cycle can look optimized purely by chance. Zorynex requires multi-cycle data, including at least one documented downturn period, before a model is classified as stable enough to present to a user.

The historical performance chart presented inside the platform plots simulated portfolio value against a benchmark index over the full backtest window, with drawdown periods shaded separately so volatility is visible rather than smoothed out.
Zorynex analysts reviewing backtesting engine output on a data terminal
Applied Use

Where Supplemental Investors Apply This

These are the three recurring scenarios reported by users managing portfolios alongside other income sources.

Case 01

Portfolio Optimization on a Limited Time Budget

Gig-economy schedules rarely allow for daily chart review. The model produces a weekly rebalancing summary so allocation decisions can be made in a single sitting instead of monitored constantly.

Case 02

Market Anomaly Detection

When price action deviates from the statistical pattern the model expects for a given asset, the position is flagged with the magnitude of the deviation, leaving the decision to act with the user.

Case 03

Strategic Rebalancing Around Irregular Income

Contribution amounts can vary month to month. The engine recalculates target allocations around each new contribution rather than assuming a fixed deposit schedule.

Transparency & Risk

Questions Worth Asking Before You Deploy Capital

No testimonials, no inflated claims — just the operating details most frequently requested during onboarding.

How transparent is the algorithm itself?

The model's feature weights and the logic behind each flagged recommendation are viewable inside the platform. Zorynex does not use undisclosed "black box" signals that cannot be traced back to a specific input.

How is data handled under German and EU regulation?

Account and portfolio data is processed on infrastructure located within the EU and handled in line with GDPR requirements. No data is sold or shared with third parties for marketing purposes.

What are the actual risks of using a model-based approach?

Backtested results describe past market behavior and do not guarantee future performance. Market conditions can shift outside the range the model was trained on, and all capital deployed carries risk of loss. Zorynex is a decision-support tool, not a guarantee of returns.

What does onboarding actually involve?

Account setup, risk-profile configuration, and your first backtest run typically take one session. No trading activity occurs automatically before you review and confirm the initial strategy parameters.

Review the Backtest Before You Decide Anything Else

Access the model's historical output, risk parameters, and data handling terms before connecting a live account.

Deploy Model Read Technical FAQ
REST API Access CSV / JSON Export PSD2 Bank Interfaces SFTP Data Feed