Use backtested AI strategies. Reduce risk and optimize entry points in real time.
Data in real time
The model processes volumes of data from multiple sources simultaneously — prices, volatility and liquidity — and generates predictive signals before a change becomes visible on a standard chart.
| An instrument | The price | A change | Volatility | Liquidity |
|---|---|---|---|---|
| EUR/USD | 1.0842 | +0.18% | Low | Tall |
| DAX40 | 18,412.50 | -0.42% | Medium | Tall |
| BTC/USD | 61,204.00 | +1.05% | Tall | Medium |
| WTI Crude | 78,32 | -0.11% | Medium | Medium |
| US10Y | 4.268% | +0.03% | Low | Tall |
Methodology
Each strategy goes through the same process before being applied to real capital: test on historical data, define risk limits and verify feasibility in multiple markets.
Strategies are tested on several years of historical data before being put into production. Results are measured through drawdown, Sharpe ratio and hit rate per market cycle.
Stop-loss parameters are set automatically based on the volatility of the instrument, not a fixed percentage. The system adjusts the exposure before the release of macroeconomic data.
The same strategies are executed in parallel on multiple markets and instruments, without manually adjusting the parameters for each pair or index.
About the platform
NovqenAltrevoAl develops models that process real-time market data and turn it into specific entry, exit and position sizing recommendations. The focus is on measurable outcomes — not on predicting market direction without evidence.
The infrastructure is designed for a low latency footprint and stable operation under load, even during periods of increased volatility when the number of signals spikes.
The results
The table shows the historical performance of the optimized NovqenAltrevoAl models against standard market indices, over the same time period and under the same market conditions.
| Strategy / Index | Annual yield | Max. drawdown | Sharpe ratio |
|---|---|---|---|
| S&P 500 (reference) | 9.80% | -23.40% | 0.68 |
| NovqenAltrevoAl — Model A | 14.20% | -11.60% | 1.24 |
| EURO STOXX 50 (reference) | 7.10% | -26.80% | 0.51 |
| NovqenAltrevoAl — Model B | 11.90% | -13.90% | 1.05 |
Risk Disclaimer: The data presented are based on backtested, historical simulation results and do not represent a guarantee of future returns. Trading in financial instruments carries the risk of losing invested capital. Past performance is not a reliable indicator of future performance.
Integration
Connecting to the existing infrastructure does not require changing the brokerage platform or terminating active positions.
Generate an API key and connect it to the broker account via the REST interface. Data is retrieved without exposing user credentials.
Define instruments, maximum exposure per position and drawdown tolerance before model activation.
The system executes the strategy according to the defined rules and records each transaction in the audit dashboard.
Questions and answers
Here we answer the most common questions about latency, security and model accuracy — without the sales pitch.
The average signal processing time is 42 ms from the receipt of the data to the generation of the recommendation. The time may vary depending on the load of the exchange and the number of active instruments.
API keys are stored encrypted and never displayed in full after generation. Access to the key is possible only through an authorized request from your own account.
Accuracy is measured through Sharpe ratio and backtest hit rate, and results vary by market and time period. No model guarantees the profitability of every single transaction.
Not. The platform connects via API with supported broker interfaces, without the need to migrate an existing account.
The system automatically reduces the size of positions and expands stop-loss limits based on the current volatility of the instrument, according to predefined risk rules.