AI-driven decision optimization
Namure Laguru converts large amounts of market data into concrete action recommendations in real time. The models are backtested against historical scenarios, so decisions are based on measurable statistics — not emotions.
Core engine
Namure Laguru's analysis engine processes price, volume and volatility data continuously and identifies patterns that are difficult to maintain manually over time. Output is not a black box — each recommendation is linked to a measurable probability and a justified risk profile.
Backtested strategy
Every strategy goes through a structured testing process where predictions are compared to actual outcomes over multiple market cycles. This reduces the risk that the model is simply adapted to the patterns of a single period.
Historical data from different volatility regimes — bull markets, corrections and sideways consolidation — are included in the test basis so that the model is not overfitted to one type of market behavior.
The model is trained on one data period and tested on a separate, unknown period. It uncovers whether the results are due to real predictive ability or random correlation in training data.
Results are reported as risk-adjusted returns — not just gross returns — so that performance can be assessed in relation to the volatility the strategy has actually been exposed to.
Risk management
The platform's risk modules are designed to limit downside before it occurs, rather than reacting after a move has already occurred.
Volatility indicators are updated per interval and is directly included in how position sizes and signal weights are adjusted.
Total portfolio risk is monitored across positions to avoid unintended concentrated exposure.
When a position moves outside defined risk limits, a notification is sent immediately.
The risk logic is built to handle both an individual trader's portfolio and a manager's overall bookkeeping, without compromising response times.
Method
Transparency in data processing is a prerequisite for trust among technical users. The process below describes the chain from data input to executable decision.
Market data, order book information and macro indicators are collected from connected sources and normalized into a common format.
Statistical and machine learning-based models identify patterns and calculate probabilities for different outcomes.
Signals are weighted against the user's risk parameters and existing portfolio composition to form a concrete recommendation.
The recommendation is made available via API or user interface, ready for manual approval or automated order placement.
About the platform
Namure Laguru is aimed at day traders, portfolio managers and fintech enthusiasts who already work data-driven, but want a systematic layer between raw data and decision. The platform is designed to complement existing research — not to replace professional assessment.
Integration takes place via a documented API, so that signals can be included in existing workflows without requiring a full platform change.
Frequently asked questions
The models use market data from connected trading platforms as well as publicly available macroeconomic indicators. Specific data providers depend on the market being traded and are configured at setup.
The platform exposes a REST-based API where signals, risk parameters and portfolio status can be pulled programmatically. Documentation and authentication method are reviewed when access is granted.
Latency depends on the update frequency of the data source and the selected market type. Signals are delivered continuously as new data is received, rather than at fixed, slower intervals.
Yes. Risk parameters, exposure limits and position sizes can be configured so that recommendations are based on the actual portfolio rather than a generic default profile.
Performance is reported on the basis of out-of-sample backtests over several market cycles, with a focus on risk-adjusted return rather than isolated gross return. The methodology is reviewed in detail when requesting access.