Vlondanaru analyzes market data in real time and dynamically adjusts stop loss levels based on volatility. Predictive models replace rigid rules with adaptive risk limits.
Request access nowStatic stop losses are based on fixed percentages or ATR multipliers. They do not take short-term regime changes into account. In volatile phases, disproportionate drawdowns occur, especially with leveraged positions.
Illustrative comparison of the functional principles. No historical performance information.
The stop loss distance is continuously recalculated based on volatility cluster data, not a rigid time interval.
Models estimate the probability of a pullback before it occurs and provide an adjusted risk limit before the price reaches it.
A classification model differentiates between trend, range and reversal phases and selects a suitable hedging logic for each phase.
Signal processing and order updates run in the millisecond range, relevant for scalping and intraday strategies.
Preview: Each position map shows the currently calculated stop distance, the underlying market regime and the timestamp of the last model adjustment.
Price, volume and order book data are recorded and normalized in parallel over several time windows.
Volatility, momentum and liquidity indicators are derived from the raw data.
A trained model estimates the probability of a drawdown occurring at the next time step.
The stop loss mark is reset and passed to the order execution layer.
The logic of the stop loss system adapts to the holding period and the execution rhythm of the respective strategy.
For positions in the second to minute range, the system reacts to micro-volatility and order book imbalances.
For multi-day positions, the model takes into account larger time windows and macroeconomic events.
Vlondanaru was designed for users who want to understand algorithmic decisions. Each stop loss level adjustment is logged and linked to the underlying model signal. The platform does not replace a trading decision, but rather provides an additional, data-driven risk limit.
The model processes tick data, order book depth and historical volatility profiles of the respective instrument. Macroeconomic event calendars are included as an additional contextual factor.
The recalculation is event-based, triggered by significant changes in volatility or liquidity, not according to a rigid time interval.
Yes. Users can define upper and lower limits within which the system is allowed to operate. Complete manual control remains possible at any time.
Currently, Vlondanaru supports liquid futures and forex pairs with sufficient historical data depth for stable model training.
Models are continually validated against out-of-sample data. Deviations between the forecast and the actual course are documented and incorporated into the next training iteration.
Technical reference for developers: Request API documentation