Skip to content
DIMECHAIN

Financial services

Portfolio OptimisationMaturity: concept

QUBO, VQE/QAOA and hybrid concepts for asset allocation and risk-return trade-offs.

concept An idea we find credible. Nothing has been built or measured.

What this is

The problem

Portfolio construction under real constraints — cardinality limits, minimum lot sizes, sector caps, turnover budgets — is a discrete optimisation problem, not the smooth one in the textbook.

Where the current approach strains

Mean-variance optimisation is elegant and continuous. Adding the discrete constraints that real mandates carry makes it combinatorial, and practitioners handle that with heuristics and manual adjustment.

What we are exploring

QUBO formulations for the constrained selection problem, where the discreteness is native to the formulation rather than bolted on.

What would have to be true

Comparison against a commercial solver on the same constraint set, and an honest note that for most portfolio sizes the commercial solver is entirely adequate. The interesting case is large universes with tight cardinality constraints, and we would need to show the boundary where that begins.

Where it applies

Related

Want to co-develop this?