Financial services
Trading OptimizationMaturity: concept
Execution optimisation, transaction costs, market impact and dynamic allocation.
concept — An idea we find credible. Nothing has been built or measured.
What this is
The problem
Executing a large order without moving the price against yourself is a scheduling problem under uncertainty: split the order across time and venues, balancing market impact against the risk that the price moves while you wait.
Where the current approach strains
Execution algorithms are mature and highly tuned. The parameters are fitted to market conditions that change. The residual is real but not obviously the kind of gap a new computational method closes.
What we are exploring
Scheduling and venue-allocation formulations. We are appropriately sceptical here — this is a domain with sophisticated incumbents and strong incentives, and "nobody thought of this" is rarely the explanation.
What would have to be true
Backtesting against realistic impact models, plus an argument for why the improvement is not already arbitraged away. That second part is the harder one.
Where it applies
Related
Insurance
Risk pricing, scenario modelling, claims/fraud analysis and portfolio management concepts.
Risk Management
Scenario analysis, stress testing, VaR and constraint optimisation.
Portfolio Optimisation
QUBO, VQE/QAOA and hybrid concepts for asset allocation and risk-return trade-offs.