Most engagements touch more than one of the areas below. A validation project usually ends in questions about execution cost, and an execution project usually starts with questions about the research behind the orders. We take the work as a whole, or one part of it, whichever the client needs.
Everything we build is data-agnostic. The same procedures run on equities, futures, currencies, options, rates and crypto assets.
Strategy validation
Formal procedures that decide whether a strategy has an edge before capital depends on it.
- Validation procedures for manual, semi-automated and automated strategies
- Robustness testing on out-of-sample data, parameter neighbourhoods and cost assumptions
- Controls for data-mining bias when many variations have been tried, including the deflated Sharpe ratio and the probability of backtest overfitting
- Independent review of existing research, backtests and code
Research infrastructure
The machinery that lets a team test ideas at scale and reproduce every result.
- Python research infrastructure, from data handling to reporting
- Distributed computing grid for strategy validation and grid execution
- Reproducible research pipelines with versioned data and results
Portfolio construction
Combining strategies so the portfolio holds up when single strategies do not.
- Decorrelation and minimum variance optimisation
- Automated retraining of evidence-based strategy portfolios as alpha decays
- Position sizing and alpha decay detection
Order and execution management
Order and execution systems that cut cost and keep flow acceptable to counterparties.
- Spread prediction and execution cost optimisation from tick and order book data
- Order aggregation and low-latency execution
- Low-latency trading infrastructure optimisation
- Automated slippage and transaction cost monitoring
- Adverse selection control, so liquidity providers and central counterparties do not flag the flow as hostile
Risk and monitoring
Measuring exposure, impact and behaviour, and acting on it by rule.
- Risk and exposure management
- Market impact and execution profile analysis
- Predictive analytics and behavioural finance systems to evaluate, hedge and manage trader-related risk
- Live monitoring that flags a strategy when it drifts away from its research
Integration and automation
Connecting brokers, venues and data, and taking manual steps out of the workflow.
- Broker, exchange and market data API integrations
- Automation of manual and semi-automated trading workflows
- Reporting for every account and strategy
How we work with clients
- Independent
- Our recommendations are not tied to any broker, venue, platform or data vendor.
- Confidential
- Work starts under a non-disclosure agreement. Client code, data and results are never reused for anyone else, and clients are not named.
- In your environment
- We can work inside your own infrastructure, so data and code never leave it.
- Negative results reported
- If a strategy fails validation, the report says so, and says why.
- Documented for review
- Every procedure is written down so your risk, compliance and audit staff can follow and repeat it.
- Clean handover
- Code, documentation and procedures are handed over in a state your team can run without us.
To discuss a project, write to our contact address.